Orientador: Alan Rodrigo Panosso / Resumo: O estudo da anisotropia espacial apresenta grande importância na modelagem e mapeamento espacial das variáveis ambientais. Assim, o objetivo do trabalho foi caracterizar a anisotropia espacial da perda de solo e atributos químicos e físicos do solo em diferentes pedoformas com cultivo de cana-de-açúcar, por meio da utilização da dimensão fractal (DF). O estudo foi realizado a partir de um banco de dados legados do Grupo de Pesquisa CSME, referente a uma área de 200 ha, localizada em Tabapuã (SP), Brasil, onde foi estabelecida uma malha de amostragem regular de 50 m, com 623 pontos. Em cada ponto, foi estimada a perda de solo por erosão (A), com a Universal Soil Loss Equation(USLE), bem como os atributos teor de argila (Arg), Areia Total (AT), Capacidade de troca de cátions (CTC), Matéria orgânica (MO) e Fósforo (P). Identificou-se a predominância de condição anisotrópica para todos os atributos na pedoforma convexa nas direções entre 90º e 150º, baseadas nos gráficos de dispersão gerados a partir de diferentes escalas e direções. Porém na pedoforma côncava A, Arg e CTC apresentaram comportamento isotrópico. Obteve-se o melhor ajuste no resultado da validação cruzada de A, na pedoforma côncava no modelo isotrópico e na pedoforma convexa no modelo anisotrópico. Concluiu-se que a pedoforma convexa proporciona maior ocorrência de anisotropia demonstrando que a forma do relevo condiciona padrões diferenciados de anisotropia espacial. / Abstract: The study of spatial anisotropy is of great importance in the modeling and spatial mapping of environmental variables. Thus, the objective of the work was to characterize the spatial anisotropy of soil loss and chemical and physical attributes of the soil in different pedoforms with sugarcane cultivation, using the fractal dimension (DF). The study was carried out from a legacy database of the CSME Research Group, referring to an area of 200 ha, located in Tabapuã (SP), Brazil, where a regular 50 m sampling grid was established, with 623 points . At each point, soil loss by erosion (A) was estimated with the Universal Soil Loss Equation (USLE), as well as the clay content (Arg), Total Sand (AT), Cation exchange capacity (CTC) attributes ), Organic matter (MO) and phosphorus (P). The predominance of anisotropic condition was identified for all attributes in the convex pedoform in the directions between 90º and 150º, based on the scatter plots generated from different scales and directions. However, in the concave pedoform A, Arg and CTC showed isotropic behavior. The best adjustment was obtained in the result of the cross-validation of A, in the concave pedoform in the isotropic model and in the convex pedoform in the anisotropic model. It was concluded that the convex pedoform provides a higher occurrence of anisotropy demonstrating that the shape of the relief conditions different patterns of spatial anisotropy. / Mestre

LOW ALTITUDE RADAR WAVE PROPAGATION MODELLING In this PhD thesis, propagation aspects of low altitude radar performance have been modeled using geometrical optics. Both the path propagation factor and the radar clutter have been modeled. Such models already exist at various complexity levels, such as round earth specular reflection combined with knife edge hill diffraction [SEKE:IEEE,Ap- 34,No:8,1980] and round earth and slant plateau reflection combined with hill diffraction [RADCAL: 1988-2000,EE,METU]. In the proposed model we have considered an extension to RADCAL&rsquo / s model to include convex and concave slant plateaus between hills and depressions (troughs). This propagation model uses a reflection model based on the Geometrical Theory of Reflection for the convex and concave surfaces. Also, back scattering from surface (clutter) is formulated for the new model of the terrain profile. The effects of the features of the terrain profile on the path propagation factor have been investigated. A real terrain data have been smoothed on the basis of the above study. In order to verify the formulation, the Divergence and Convergence Factors associated with the convex and concave plateaus, respectively are inserted into the RADCAL program. The chosen terrains have convex or concave plateaus in the model. The output of the RADCAL is compared with measured values and other propagation algorithms such as Forward-Backward Spectrally Accelerated (FBSA) [FBSA:IEEE Vol.53, No:9,2005] and Parabolic Equation Method [TPEM:IEEE Vol.42,No:1,1994]. Moreover, as the RADCAL Propagation model is based on the ray optics, the results are also compared with another ray optics based propagation model. For this purpose the results of SEKE [Lincoln Lab.] propagation model are used. SEKE model has been used to compute path loss for different types of terrain as a function of receiving antenna height at a fixed distance between transmit and receive antennas. For Beiseker W35 Terrain profile, the results of RADCAL, SEKE and measurements are compared. All results are in good agreement with those of RADCAL.

Abstract The influence of rotating and cross flow effect on local heat transfer coefficient and flow visualization for a single confined air/water jet with jet-to-wall spacing from 5 to 11.4, jet Reynolds number from 6500 to 26000, rotational Reynolds number from 0 to 112000, curvature ratio from 150 to , ratio of crossflow massflux to jet mass flux from 0 to 2, and the heat flux from 1430 to 12890W/m2 were reported. The local heat transfer coefficient for air/water along the surface is measured and the effect of the rotation, the jet-to-wall spacing, the surface curvature, local and average Nusselt number, are presented and discussed. Furthermore, flow visualization was made in the present study. Based on the experimental result, it is found that the rotation will induce the centrifugal and coriolis force. It also shows that the heat transfer response will be decreased when the impinging direction parallel to the rotating direction, and increased when impinging direction perpendicular to the rotating direction. Crossflow effect will make Nusselt number decrease to 48% when M=2. Moreover, the roughen surface will increase the heat transfer coefficient up to 22% due to the secondary flow. The flow visualization is used to observe the transition of laminar to turbulence flow and to calculate the boundary layer thickness.