Abstract

Abstract

Abstract

Abstract
This paper reports the design of a new remotely operated underwater vehicle (ROV), which has been developed at the Underwater Systems and Technology Laboratory (USTL) - University of Porto. This design is contextualized on the KOS project (Kits for underwater operations). The main issues addressed here concern directional drag minimization, symmetry, optimized thruster positioning, stability and layout of ROV components. This design is aimed at optimizing ROV performance for a set of different operational scenarios. This is achieved through modular configurations which are optimized for each different scenario.

Abstract
The need to generate random wave fields with a pre-determined spectrum is a common one. One of the typical applications requiring this capability is the need to interface spectral and deterministic models. The spectra and integral parameters produced as output of spectral models must often be transformed into wave fields which can be used as input to deterministic models (e.g. Navier-Stokes based numerical models). Also, the instantaneous statistics of the produced wave field must be controlled. This is typically the case when the objective is to evaluate the influence of these statistics in the appearance of certain wave phenomena. In this article we propose a parametric based approach to the wave field generation. Not only is this a computationally efficient way of generating local elevation sequences, but it also enables perfect control of both the instantaneous and spectral characteristics of the generated waves. Two examples of applications will be given. One, where the developed generator will be used to investigate the influence of the spectral model and the instantaneous statistics of the wave field in extreme wave generation. A second example will be the generation of a boundary field for a Navier-Stokes based model starting from the spectral output of the SWAM model.

Abstract
Spatial and temporal sea level variability in the North Atlantic is investigated from Topex/Poseidon (T/P) altimetry data. Time series of sea level anomalies on a regular 5 degrees grid are analysed. Non-linear denoising through thresholding in the wavelet transform domain is carried out for each series in order to remove noise while preserving non-smooth features. Principal Component Analysis (PCA) is used to obtain a spatio-temporal description of the sea level field, To avoid modal mixing and improve interpretation of the principal modes, PCA is implemented separately for seasonal and trend components of the sea level field obtained from a wavelet-based multiresolution analysis. The leading pattern of the seasonal field reflects the dominance of a stable annual cycle over the study area and the change in the seasonal regime approaching the equator with contribution of the semi-annual cycle and phase-shift in the annual cycle in the tropical Atlantic. The leading pattern of the trend field is a broad spatial pattern associated with North Atlantic Oscillation (NAO), reflecting the influence of atmospheric conditions on interannual sea level variability.