Mathematical modeling of population dynamics can provide novel insight to the growth and dispersal patterns for a variety of species populations, and has become vital to the preservation of biodiversity on a global-scale. These growth and dispersal stages can be modeled using integrodifference equations that are discrete in time and continuous in space. Previous studies have identified metrics that can determine whether a given species will persist or go extinct under certain model parameters. However, a need for computational tools to compute these metrics has limited the scope and analysis within many of these studies. We aim to create computational tools that facilitate numerical explorations for a number of associated integrodifference equations, allowing modelers to explore results using a selection of models under a robust parameter set.

Orientador: Wilson Castro Ferreira Junior / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matematica, Estatistica e Computação Cientifica / Made available in DSpace on 2018-08-12T08:00:18Z (GMT). No. of bitstreams: 1 LouVega_Salvador_M.pdf: 1338284 bytes, checksum: f247d3dc13a783da27d224a3d32df47b (MD5) Previous issue date: 2008 / Resumo: Proponemos um modelo matemático para uma planta invasora, que acopla a dinâmica de reproduão com Efeito Allee e a dispersão de longa distância de uma planta invasora. Consideramos um efeito Allee devido à limitação de pólen, que reduz a produção de sementes. Introduzimos o efeito Allee através de uma função de probabilidade de encontro pólen-estigma que depende da densidade de plantas. Para a modelagem do processo de dispersão utilizamos equações íntegrorecursiva (IRE) tomando um núcleo de dispersão misto, que representa a dispesão local e a longa distância. Analisamos a dinâmica local do modelo determinando os pontos de equilíbrio e as suas estabilidades, para então analisar o processo de dispersão. Analisamos o modelo de dispersão por meio de simulação numérica, o que permitiu observar o deslocamento espacial da frente da invasão. Isto permitiu calcular a velocidade de expansão. Determinamos a inuência do efeito Allee, da capacidade reprodutiva e da dispersão de longa distância sobre a velocidade de expansão. Observamos que o efeito Allee torna velocidades aceleradas em velocidades constantes de expansão. A velocidade de expansão decresce com o aumento na intensidade do efeito Allee, mas aumenta com a capacidade reprodutiva. A dispersão de longa distância gera maiores velocidades de expansão, embora para fortes intensidades do efeito Allee o acréscimo na velocidade não é signifícativo em relação à velocidade gerada pela dispersão local. Os resultados mostram que apesar da dispersão contribuir ao aumento na velocidade de expansão, a dispersão também torna a população mais suscetável á extinção. / Abstract: We present a mathematical model which couples the reproductive dynamic with an Allee effect and a long distance diseprsal of an invasive plant. We consider an Allee effect due to pollen limitation, which reduces seed production. We introduce the Allee effect through a probability function that describes pollen-stigma encounters as function of the population density. To model the dispersal process we used integro-diference equations (IDE) and employed a mixed kernel which represents the local and long distance dispersal processes. We analyzed the local dynamic through the stability of their equilibrium points. For the spatial dynamic we used numerical simulations, that allowed us to observe the spatial displacement of the invasion front. This permitted us to compute the expansion speeds. We determined the inuence of the Allee effect, reproductive capacity and the long distance diseprsal on the invasion speeds. We observed than an Allee effect turns accelerating expansion speeds into constant speeds. Expansion speeds decreases with Allee effect intensity but increases with the reproductive capacity of the population. Long distance dispersal produces higher invasion speeds, but for strong intensities of the Allee effect, the increase is not significant in relation to the speeds generated by the local dispersal. Our results show that while dispersal contributes to expansion speeds, it also turns the population more susceptible to extinction. / Mestrado / Biomatematica / Mestre em Matemática Aplicada

No description available.

La dispersion du pollen est une composante majeure des flux de gènes chez les plantes, contribuant à la diversité génétique et à sa structure spatiale. Son étude à l'échelle d'un épisode de reproduction permet de comprendre l'impact des changements actuels (fragmentation, anthropisation....) et de proposer des politiques de conservation. Deux types de méthodes basées sur les marqueurs microsatellites estiment la fonction de dispersion du pollen: (i) les méthodes directes (e.g. mating model) basées sur l'assignation de paternité et nécessitant un échantillonnage exhaustif (position et génotype des individus du site étudié, génotypes de graines échantillonnées sur des mères); (ii) les méthodes indirectes (e.g. TwoGener), nécessitant un échantillonnage réduit (génotypes des graines, génotypes et positions des mères) et résumant les données en indices génétiques. Nous proposons la formalisation statistique de ces deux types de méthodes et montrons qu'elles utilisent des fonctions de dispersion différentes: les méthodes directes estiment une fonction forward potentielle (déplacement du pollen depuis le père), les méthodes indirectes une fonction backward intégrative (de la fécondation jusqu'à l'existence du père). Nous explicitons le lien entre fonctions backward et forward, des hypothèses menant à leur équivalence, et des contraintes affectant les fonctions backward. Nous développons enfin une méthode de calcul bayésien approché qui permet (i) une estimation forward, (ii) avec des intervalles de crédibilité, (iii) à partir d'un jeu de données non exhaustif et d'informations partielles (e.g. positions sans génotype) et (iv) l'utilisation de différents modèles de dispersion. / Pollen dispersal is a major component of gene flow in plants. It determines to genetic diversity and spatial genetic structure.Studying it at the scale of a single reproduction event enables to understand the impact of current changes (fragmentation, anthropization ...) and to propose conservation practices.Two types of methods, based on microsatellite markers, estimate pollen dispersal functions : (i) direct methods (e.g. mating model) based on paternity assignment require exhaustif sampling (position and genotype of individuals in the study plot, genotypes of seeds harvested on mothers); (ii) indirect methods (e.g. TwoGener), require a weaker sampling (seeds genotypes, genotypes and positions of their mothers) and summarize data through genetic indices.We propose a statistical formalization of both types of methods and show that they rely on different dispersal functions : the direct methods estimate a potential forward function (pollen transfer from the father), whereas the indirect methods estimate an integrative backward one (from fecondation to father existence). We exhibit the link between forward and backward functions, assumptions leading to their equivalence and constrains affecting the backward functions.Finally, we develop an Approximate Bayesian Computation method, which enable (i) a forward estimation, (ii) with credibility intervals, (iii) from a non exhaustive dataset and partial information (e.g. positions without genotypes) and (iv) the use of different dispersal models.

The main goal of this study was to evaluate the effectiveness of two biological control agents in reducing population growth and spread of the invasive tree Melaleuca quinquenervia, a subtropical tree native to Australia, and invasive in Florida, Puerto Rico, and the Bahamas. While in Florida two insects Oxyops vitiosa (weevil), and Boreioglycaspis melaleucae (psyllid) have been established as biocontrol agents, in Puerto Rico only psyllids are present, and in the Bahamas no biocontrol agents are present. This study combined demographic data, experiments and mathematical models to investigate the influence of the biocontrol agents on M. quinquenervia's spatial population dynamics. In the field, permanent plots were established and demographic data was collected in populations in the native and exotic ranges. Australian populations are comprised mostly of tall adult trees, while in the exotic ranges populations are comprised mostly of short trees (<1.3m in height), and small adult trees. In a shade-house, I performed an experiment to investigate the effects of insect type and density on survival and growth of M. quinquenervia seedlings. I found that high density of insects, independently or in combination, reduce seedling performance, thus having the potential to alter the seedling-short plant transition of the M. quinquenervia life cycle. Based on the demographic data, I developed integral projection models (IPMs) to determine population growth rates in each region. Populations in Australia and the Bahamas are increasing, while populations in Florida and Puerto Rico are decreasing. Population growth is most sensitive to the seedling-short plant transition in all regions, except Florida, where it is most sensitive to survival of tall plants. Simulations combining the results of the IMPs and experiment indicated the biocontrol damage results in reductions in population growth rate in Puerto Rico and the Bahamas. Seed dispersal and demographic data was combined to develop an integrodifference structured model of population spread. Simulations indicated that by reducing seedling performance, insects have the potential to reduce the rate of population spread. Overall this study shows that individual-level effects of biocontrol agents have translated into reductions in population growth rate and rate of spread of M. quinquenervia.

In this dissertation, we study integrodifference equations in patchy landscapes. Specifically, we provide a framework for linking individual dispersal behavior with population-level dynamics in patchy landscapes by integrating recent advances in modeling dispersal into an integrodifference equation. First, we formulate a random-walk model in a patchy landscape with patch-dependent diffusion, settling, and mortality rates. We incorporate mechanisms for individual behavior at an interface which, in general, results in the probability-density function of the random walker being discontinuous at an interface. We show that the dispersal kernel can be characterized as the Green's function of a second-order differential operator and illustrate the kind of (discontinuous) dispersal kernels that arise from our approach. We examine the dependence of obtained kernels on model parameters. Secondly, we analyze integrodifference equations in patchy landscapes equipped with discontinuous kernels. We obtain explicit formulae for the critical-domain-size problem, as well as, explicit formulae for the analogous critical size of good patches on an infinite, periodic, patchy landscape. We examine the dependence of obtained formulae on individual behavior at an interface. Through numerical simulations, we observe that, if the population can persist on an infinite, periodic, patchy landscape, its spatial profile can evolve into a discontinuous traveling periodic wave. We derive a dispersion relation for the speed of the wave and illustrate how interface behavior affects invasion speeds. Lastly, we develop a strategic model for the spread of the emerald ash borer and its interaction with host trees. A thorough literature search provides point estimates and interval ranges for model parameters. Numerical simulations show that the spatial profile of an emerald ash borer invasion evolves into a pulse-like solution that moves with constant speed. We employ Latin hypercube sampling to obtain a plausible collection of parameter values and use a sensitivity analysis technique, partial rank correlation coefficients, to identify model parameters that have the greatest influence on obtained speeds. We illustrate the applicability of our framework by exploring the effectiveness of barrier zones on slowing the spread of the emerald ash borer invasion.

Dispersal, especially long-distance dispersal, is an important component in many disciplines within biology. Many species are passively dispersed by wind, not least spore-dispersed organisms. In this thesis I investigated the dispersal capacity of bryophytes by studying the colonization patterns from local scales (100 m) to landscape scales (20 km). The dispersal distances were measured from a known source (up to 600 m away) or inferred from a connectivity measure (1–20 km). I introduced acidic clay to measure the colonization rates over one season of a pioneer moss, Discelium nudum (I–III). I also investigated which vascular plants and bryophytes that had colonized limed mires approximately 20–30 years after the first disturbance (IV). Discelium effectively colonized new disturbed substrates over one season. Most spores were deposited up to 50 meters from a source but the relationship between local colonization rates and connectivity increased with distance up to 20 km (I–III). Also calcicolous wetland bryophyte species were good colonizers over similar distances, while vascular plants in the same environment colonized less frequently. Common bryophytes that produce spores frequently were more effective colonizers, while no effect of spore size was detected (IV). A mechanistic model that take into account meteorological parameters to simulate the trajectories for spores of Discelium nudum fitted rather well to the observed colonization pattern, especially if spore release thresholds in wind variation and humidity were accounted for (III). This thesis conclude that bryophytes in open habitats can disperse effectively across landscapes given that the regional spore source is large enough (i.e. are common in the region and produce spores abundantly). For spore-dispersed organisms in open landscapes I suggest that it is often the colonization phase and not the transport that is the main bottle-neck for maintaining populations across landscapes. / <p>At the time of the doctoral defence the following papesr were unpublished and had  a status as follows: Paper 2: Epubl ahead of print; Paper 3: Manuscript; Paper 4: Manuscript</p>

In this dissertation, we study integrodifference equations in patchy landscapes. Specifically, we provide a framework for linking individual dispersal behavior with population-level dynamics in patchy landscapes by integrating recent advances in modeling dispersal into an integrodifference equation. First, we formulate a random-walk model in a patchy landscape with patch-dependent diffusion, settling, and mortality rates. We incorporate mechanisms for individual behavior at an interface which, in general, results in the probability-density function of the random walker being discontinuous at an interface. We show that the dispersal kernel can be characterized as the Green's function of a second-order differential operator and illustrate the kind of (discontinuous) dispersal kernels that arise from our approach. We examine the dependence of obtained kernels on model parameters. Secondly, we analyze integrodifference equations in patchy landscapes equipped with discontinuous kernels. We obtain explicit formulae for the critical-domain-size problem, as well as, explicit formulae for the analogous critical size of good patches on an infinite, periodic, patchy landscape. We examine the dependence of obtained formulae on individual behavior at an interface. Through numerical simulations, we observe that, if the population can persist on an infinite, periodic, patchy landscape, its spatial profile can evolve into a discontinuous traveling periodic wave. We derive a dispersion relation for the speed of the wave and illustrate how interface behavior affects invasion speeds. Lastly, we develop a strategic model for the spread of the emerald ash borer and its interaction with host trees. A thorough literature search provides point estimates and interval ranges for model parameters. Numerical simulations show that the spatial profile of an emerald ash borer invasion evolves into a pulse-like solution that moves with constant speed. We employ Latin hypercube sampling to obtain a plausible collection of parameter values and use a sensitivity analysis technique, partial rank correlation coefficients, to identify model parameters that have the greatest influence on obtained speeds. We illustrate the applicability of our framework by exploring the effectiveness of barrier zones on slowing the spread of the emerald ash borer invasion.