Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-06-06T13:21:41Z No. of bitstreams: 1 joaovitordesahauck.pdf: 33226275 bytes, checksum: b7a8ed469369e9bbccd22738ced13ab0 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-06-06T14:07:12Z (GMT) No. of bitstreams: 1 joaovitordesahauck.pdf: 33226275 bytes, checksum: b7a8ed469369e9bbccd22738ced13ab0 (MD5) / Made available in DSpace on 2017-06-06T14:07:12Z (GMT). No. of bitstreams: 1 joaovitordesahauck.pdf: 33226275 bytes, checksum: b7a8ed469369e9bbccd22738ced13ab0 (MD5) Previous issue date: 2015 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Malhas com restrições no comprimento das arestas são úteis para diversas aplicações, especialmente para simulações de processos químicos e físicos. Este trabalho apresenta um método iterativo para remalhar uma malha triangular arbitraria de variedade 2 em uma malha com o comprimento de todas as arestas dentro de um intervalo de restrição definido pelo usuário. O método usa operações estelares para ajustar a quantidade de vértices e triângulos no modelo e para melhorar a valência dos vértices. Ele também aplica o operador Laplaciano em um espaço paramétrico local para melhorar a distribuição de vértices sobre a superfície. Propõe-se, uma otimização não linear, aplicada localmente, para os casos em que a malha é praticamente regular. Perdas geométricas são evitadas pela realização de uma projeção sobre a superfície original. O método proposto resulta em uma malha praticamente regular, com os vértices distribuídos uniformemente sobre a superfície. A dual da malha é usada em simulações de nano estruturas de carbono como uma aplicação do método. A principal contribuição deste trabalho é uma nova abordagem para restringir explicitamente o comprimento das arestas em um intervalo dado. Nosso método ainda garante baixa perda global de geometria e baixo custo de memória em comparação com métodos disponíveis na literatura. / Meshes with constraints in the edge length are useful for several applications, spe-cially for chemical and physical simulations. This work presents an iterative method for remeshing an arbitrary triangular 2-manifold mesh into a mesh with all edge lengths within an user-defined constraining interval. The method uses stellar operations to adjust the amount of vertices and triangles in the model and for improving the valence of the vertices. It also applies the Laplacian operator in a local parametric space to improve the distribution of the vertices over the surface. We propose a nonlinear optimization, locally applied, for cases in which the mesh is almost regular. Geometric losses are prevented by performing a projection over the original surface. Our method results in a nearly regu-lar mesh, with vertices uniformly distributed over the surface. The dual of the mesh is used in simulations of carbon nanostructures as an application of the method. The main contribution of this work is a new approach for constraining the edge length within an explicitly given interval. Our method also ensures lower global geometry losses and lower memory cost in comparison to methods available in the literature.

This thesis considers two distinct topics in phylogenetic analysis. The first is construction of empirical rate matrices for amino acid models. The second topic, which constitutes the majority of the thesis, involves analysis of and extensions to the BH model of Barry and Hartigan (1987). There are a number of rate matrices used for phylogenetic analysis including the PAM (Dayhoff et al. 1979), JTT (Jones et al. 1992) and WAG (Whelan and Goldman 2001). The construction of each of these has difficulties. To avoid adjusting for multiple substitutions, the PAM and JTT matrices were constructed using only a subset of the data consisting of closely related species. The WAG model used an incomplete maximum likelihood estimation to reduce computational cost. We develop a modification of the pairwise methods first described in Arvestad and Bruno that better adjusts for some of the sparseness difficulties that arise with amino acid data. The BH model is very flexible, allowing separate discrete-time Markov processes to occur along different edges. We show, however, that an identifiability problem arises for the BH model making it difficult to estimate character state frequencies at internal nodes. To obtain such frequencies and edge-lengths for BH model fits, we define a nonstationary GTR (NSGTR) model along an edge, and find the NSGTR model that best approximates the fitted BH model. The NSGTR model is slightly more restrictive but allows for estimation of internal node frequencies and interpretable edge lengths. While adjusting for rates-across-sites variation is now common practice in phylogenetic analyses, it is widely recognized that in reality evolutionary processes can change over both sites and lineages. As an adjustment for this, we introduce a BH mixture model that not only allows completely different models along edges of a topology, but also allows for different site classes whose evolutionary dynamics can take any form.

The boreal forest in Fennoscandia has been subjected to major loss and fragmentation of natural forests due to intensive forestry. This has resulted in that forest edges are now abundant and important landscape features. Edges have documented effects on the structure, function and biodiversity in forests. Edge influence on biodiversity is complex and depends on interactions between many local and regional factors. This thesis focuses on sharp forest edges and their potential to influence biodiversity at the landscape-level. I have developed a method for quantification and characterization of sharp forest edges by interpretation of colour infrared (CIR) aerial photographs in combination with line intersect sampling (LIS) and sample plots. The method was used to estimate density of forest edge in 28 landscapes (each 1600 ha) in northern Sweden, differing in management intensity, landscape composition and geographical location. Forest edges were described in detail using edge, canopy and neighbourhood attributes. By combining these attributes it was possible to classify edges with respect to levels of exposure. A field experiment was conducted to examine the effect of edge contrast on growth of the old forest lichen Usnea longissima. The edge quantification method is accurate and efficient for estimating the length of sharp forest edges on an area basis (edge density, m ha-1) and for collecting detailed attributes of edges and their surroundings. In northern Sweden, the forest edge density is high (54 m ha-1) but varies extensively (12-102 m ha-1) between landscapes. Edge density is strongly correlated with the level of human disturbance and increases towards the southern part of the study area, at lower altitudes were management intensity is highest. Edge orientation, contrast and neighbourhood size shows an immense variation between edges and also varies between edge types. Regenerating edges are generally of higher contrast and face larger neighbourhoods than natural edges. Maintained edges had high contrast but small neighbourhoods. A larger proportion of edges in mature forests are highly exposed to microclimatic edge influence than edges in general. The field experiment revealed that growth of U. longissima was highest near edges where the vegetation on the adjacent area was sheltering, but not shading, the lichen. In the present thesis, I have provided a valuable tool for estimating density of forest edges with potential to yield information on important factors determining edge influence at landscape-level. The large variability in edge density, edge and neighbourhood attributes imply large differences in microclimate anf thus in the potential for ede influence. Management and conservation strategies must incorporate these factors to realistically address edge influence on biota at the landscape-level.