Abstract
Ocean sewage outfalls are major sources of contaminants to coastal ocean ecosystems. This method of disposal has advantages in terms of economy and relative societal impact, but it also raises important concerns about public health and ecosystem preservation. Autonomous Underwater Vehicles have already been shown to be very useful for monitoring routine of ocean outfalls. The major advantage of this technology over traditional methods is the ability to collect high-resolution data which can be very valuable for environmental impact assessment and comparison with plume prediction models. Once the data has been collected in the field it is necessary to extrapolate from monitoring samples to unsampled locations. Geostatistics has been successfully used to obtain information, for example, regarding the spatial distribution of soil properties. In this work geostatistics is used to model and map the spatial distribution of temperature and salinity measurements gathered by MARES AUV in a monitoring campaign to Foz do Arelho outfall, with the aim of distinguishing the effluent plume from the receiving waters and characterizing its spatial variability in the vicinity of the discharge. The results demonstrate that this methodology provides good estimates of the dispersion of effluent and it is therefore very valuable in assessing the environmental impact and managing sea outfalls.

Abstract
Several monitoring approaches have been used to understand the physical, chemical, and biological processes associated with coastal sewage discharges. However, these efforts have not improved the understanding of the interaction of effluent plume/coastal ocean processes. Autonomous underwater vehicles (AUVs) have already been shown to be very useful for performing high-resolution surveys of small features such as outfall plumes. Some of the advantages of these platforms include easier field logistics, low cost per deployment, good spatial coverage, sampling over repeated sections, and the ability to perform feature based or adaptive sampling. Once the data have been collected in the field, it is necessary to extrapolate from monitoring samples to unsampled locations. Geostatistics has been successfully used to obtain information; for example, regarding the spatial distribution of soil properties. Besides giving estimated values at unsampled locations, it provides a measure of the accuracy of the estimate, which is a significant advantage over traditional methods used to assess pollution. In this work, geostatistics is used to model and map the spatial distribution of temperature measurements gathered by an AUV in a sea ouffall monitoring campaign, with the aim of distinguishing the effluent plume from the receiving waters and characterizing its spatial variability in the vicinity of the discharge. The results demonstrate that this methodology can provide good estimates of the dispersion of effluent, and it is therefore very valuable in assessing the environmental impact and managing sea outfalls.

Abstract
The wastewater plumes show to be very difficult to observed in detail. The several studies already conducted exhibit very complex and patchy structures both in vertical and horizontal sections. It is not clear if this plume patchiness is due to physical processes or measurement limitations. Rapid tow-yo sampling is expected to reduce the time variability during and between transects. The AUVs may be a useful instrument to map and detect wastewater plumes. This paper presents several prediction studies using time series files of actual in-situ measurements integrated in a near field model. The model predictions of the plume characteristics at the end of near field support the definition of the best sampling strategy for an AUV monitoring mission in a Portuguese west coast outfall.

Abstract
Near field model predictions can be used to reduce the uncertainty about outfall sewage plume location during a monitoring mission. In this paper we present an innovative methodology that uses these models to predict the plume location, establishing a sensing strategy applicable to a monitoring mission using an AUV. The paper describes two applications that implement this methodology, accomplishing the automatic mission definition using real-time oceanographic data.

Abstract

Abstract
In this article, we present the design and development of a localisation system for a mobile autonomous platform. The purpose of this system is to endow this vehicle with the capability of maintaining a position estimate in real time without artificial aids in a structured environment. Each position estimate is generated from the previous one with information obtained from the on board sensors as well as from the world model. Process and observation models are built in such a way that take into account the physical restrictions of the vehicle and its ultra-sonic sonars. All the stages of the Extended Kalman Filtering (EKF) process applied to the vehicle of interest are described. To increase the performance of the sonar data real time acquisition, a method based on entropy choosing the sonar observation in order to minimise the uncertainty of the position estimate is used. An adaptive process of matching the sonar observation decreasing the possibility of loss of the vehicle is also chosen. Satisfactory results showing the good performance of this localisation system are presented.