Abstract

Abstract
A monitoring mission to study the shape and estimate initial dilution of the S. Jacinto outfall plume using an autonomous underwater vehicle (AUV) was performed on July 30, 2002. In order to reduce the uncertainty about plume location and to concentrate the vehicle mission only in the hydrodynamic mixing zone, outputs of a near-field prediction model, based on effective real-time in situ measurements of current speed and direction and density stratification, were opportunistically used to specify in real time the mission transects. The surface characteristics of the outfall plume were found to be influenced strongly by the relatively weak stratification and low current velocities. Dilution was estimated using a tempera tu re-salinity (TS-) diagram with initial mixing lines between wastewater and ambient waters. Effluent dilutions were at least 30:1 in this study. In order to efficiently map the plume dispersion we applied the least-squares collocation method technique. Our results demonstrate that AUVs can provide high-quality measurements of physical properties of effluent plumes in a quite effective manner and valuable considerations about the initial mixing processes under real oceanic conditions can be further investigated.

Abstract
In this work a methodology is developed of optimization applying to the Simulated Annealing Algorithm, which involves the dimensioning and the localization of STATCMOs at the different buses of the Portuguese grid where there is injection of the wind generation considering the 2009 scenario. With this methodology it is hoped to obtain reactive energy support and consequently improving the voltage values at situations of abnormal grid functioning, as for exemple, at the sequence of a short-circuit In this way it is supplied, to the wind generators capacity of "ride through fault" capability so that the Grid Codes are fulfilled defined by the transmission system operators. The results show how important can be STACOMS contributions in case of failure. Avoiding in many situations that the wind generators protections act due to the regulation of the relays of under-voltage, contributing in this way for the improvement of the dynamic behavior of the whole grid.

Abstract
We derive likelihood ratio tests for the equality of the directional parameters of k bipolar Watson distributions defined on the hypersphere with common concentration parameter. We analyze the power of these tests in the case of two distributions supposing in the alternative hypothesis two directional parameters forming an angle, which varies from 18 degrees to 90 degrees. We also compare the likelihood ratio tests with a high-concentration F-test.

Abstract
In this paper we investigate how the complex rotation and quivering motion of an elongated polarized dust grain in the presence of a monochromatic electromagnetic (EM) wave can produce dipolar emission with two distinct spectral components. We present a model for the emission of radiation by elongated polarized dust grains under the influence of both an external EM wave and a constant background magnetic field. The dust, exhibiting rotational motion at the external EM field frequency w(0) as well as quivering motion at a frequency Omega(0), proportional to the EM field amplitude, will radiate with frequencies that will depend on the external field wavelength and amplitude. The radiated spectra exhibits a frequency around w(0), and sidebands at w(0) +/- Omega(0) and w(0) +/- 2 Omega(0). Since the amplitude and the frequency of the background EM field are independent parameters, this model establishes a correlation between different spectral components of galactic dipolar emission, which may help to explain the correlation between a component of the Galactic microwave emission and the 100 mu m thermal emission from interstellar dust that has been recently measured.

Abstract
A Bayesian segmentation approach for hyperspectral images is introduced in this paper. The method improves the classification performance of discriminative classifiers by adding contextual information in the form of spatial dependencies. The technique herein presented builds the class densities based on Fast Sparse Multinomial Logistic Regression and enforces spacial continuity by adopting a Multi-Level Logistic Markov-Gibs prior. State-of-art performance of the proposed approach is illustrated in a set of experimental comparisons with recently introduced hyperspectral classification/segmentation methods.