Abstract
The maintenance of plants in the increasing number of gardens and in the revegetation of degraded landscape areas generates a rising demand of water. However, shortage of water due to low and irregular rainfalls in semiarid zones such as the Mediterranean requires a rationalised use of water. In view of this, measure reducing water consumption such as minimum irrigation technique together with the use of native species with the ability to maintain its minimum ornamental value under stress watering conditions are essential for low water consumption. The aim of this work was to determine the minimum watering dosage in two native Mediterranean species (Santolina chamaecyparissus L. and Arbutus unedo L.) in order to maintain its ornamental value for their use in green areas. The assay was carried out in Northeast of Spain. Plants were distributed in Is independent plots sized 1.1 m(2). One third of the plants (6 plots) were not irrigated whereas the remaining two thirds received weekly through drip irrigation the quantity of water corresponding to 20% and 40% of the ETP0, respectively. During this study were regularly evaluated soil water content, atmospheric environmental conditions, survival rate, growth in height and surface (covering of the ground), intensity of the green colour and blossoming's intensity and longevity of Santolina chamaecyparissus L. flowers. Both species show significant difference in studied parameters between them. The 40% treatment compared to 0% treatment promoted significant differences in al studied parameters in Santolina chamaecyparissus L. and only in height and SPAD in Arbutus unedo L. The 20% irrigation treatment compared to the 0% treatment only promoted significant differences in SPAD of both species. The irrigation treatments also promoted that differences in growth rate started before than in non-irrigated treatment. From these results we can conclude that reduced water applications can improve ornamental value of xerigardening and landscaping plants under Mediterranean conditions.

Abstract
Naive Bayes is a well known and studied algorithm both in statistics and machine learning. Although its limitations with respect to expressive power, this procedure has a surprisingly good performance in a wide variety of domains, including many where there are clear dependencies between attributes. In this paper we address its main perceived limitation - its inability to deal with attribute dependencies. We present Linear Bayes that uses, for the continuous attributes, a multivariate normal distribution to compute the require probabilities. In this way, the interdependencies between the continuous attributes are considered. On the empirical evaluation, we compare Linear Bayes against a naive-Bayes that discretize continuous attributes, a naive-Bayes that assumes a univariate Gaussian for continuous attributes, and a standard Linear discriminant function. We show that Linear Bayes is a plausible algorithm, that competes quite well against other well established techniques.

Abstract
The aim of this study is the application of ERS data (ATSR2 and radar altimetry) to fisheries monitoring in the Cape Verde region. This work is being done in the scope of the ESA A03-265 project. The Cape Verde Archipelago lies under the influence of the southern portion of the Canarias Current, where it turns westward and becomes the North Equatorial Current. Cold and nutrient-rich waters upwelled at the Western African coast are transported south-westward along this current system. However, due to the seasonal north-south migration of the current system, the south-eastern region of the islands can be affected by the Equatorial Countercurrent. Two very different water masses meet forming a large scale frontal system in the vicinity of the archipelago. These frontal zones are potentially favourable for the aggregation of large pelagic highly migratory fish species. The knowledge of the front location and variability has a great socio-economic importance for the local fishing industry, as well as in the support of global stock management of these species. Using maps of sea surface topography and sea level anomalies derived from satellite altimetry and maps of sea surface temperature from ATSR2, the general features of the ocean circulation in the study area have been investigated. Due to the fact that Cape Verde region has very cloudy conditions, the synergy of microwave and infra-red data is of great importance for this study. The methodology used in this analysis and the results obtained so far are presented. Future work involves the use of in situ data to validate satellite derived information and the investigation of the relationships between fishing success and the ocean conditions measured by satellite and in situ methods. The ultimate goal will be the development of an operational methodology for the location of potential favourable areas for the concentration of large pelagic fish in the area of local fishing fleet activity.

Abstract
A robust feedback control of an induction motor is proposed. The theoretical basis is a continuous-time machine model defined by a pair of simultaneous complex-coefficients differential equations. The parameter uncertainties and load perturbations are considered in the controller design. The control methodology is based on sliding mode control techniques. The design of a sliding mode controller for tracking the torque and the rotor flux of an induction motor drive is also given. The design includes the hardware structure, control laws for torque control and rotor flux control. The modulation is defined as part of the variable structure control algorithm which enables an optimum choice of the output voltage of the static converter. The validity and robustness of the proposed control system were verified by simulation. The simulation results are presented to demonstrate their agreement with the theoretical predictions of the proposed controller. Copyright (C) 2000 John Wiley & Sons, Ltd.

Abstract
In this paper, an advanced control system for the optimal operation and management of isolated power systems with increased renewable power integration is presented. The control system minimises the production costs through on-line optimal scheduling of the power units, taking into account short-term forecasts of the load and the renewable resources. The power system security is supervised via on-line security assessment modules, which emulate the power system frequency changes caused by pre-selected disturbances. For each of the above functions, a number of techniques have been applied, both conventional and AI based. The system has been installed in the dispatch center of Crete since June 1999, and is under evaluation. © 2000 IEEE.

Abstract
A compact all-fiber interrogation technique for Bragg sensors is presented. It is based on the generation of an electric carrier by modulating the coupling length of a wavelength-division multiplexer (WDM). Theoretical and experimental results are presented when both serrodyne and sinusoidal modulations are considered. It is shown that this technique provides static microstrain resolution in measurement intervals larger than some millistrain.