Abstract
This paper presents the development of a bivalve farmer agent interacting with a realistic ecological simulation system. The purpose of the farmer agent is to determine the best combinations of bivalve seeding areas in a large region, maximizing the production without exceeding the total allowed seeding area. A system based on simulated annealing, tabu search, genetic algorithms and reinforcement learning, was developed to minimize the number of iterations required to unravel a semi-optimum solution by using customizable tactics. The farmer agent is part of a multi-agent system where several agents, representing human interaction with the coastal ecosystems, communicate with a realistic simulator developed especially for aquatic ecological simulations. The experiments performed revealed promising results in the field of optimization techniques and multi-agent systems applied to ecological simulations. The results obtained open many other possible uses of the simulation architecture with applications in industrial and ecological management problems, towards sustainable development.

Abstract
Simulating ecological models is always a difficult task, not only because of its complexity but also due to the slowness associated with each simulation run as more variables and processes are incorporated into the complex ecosystem model. The computational overhead becomes a very important limitation for model calibration and scenario analysis, due to the large number of model runs generally required. This paper presents a framework for ecological simulations that intends to increase system performance through the ability to do parallel simulations, allowing the joint analysis of different scenarios. This framework evolved from the usage of one simulator and several agents, that configure the simulator to run specific scenarios, related to possible ecosystem management options, one at a time, to the use of several simulators, each one simulating a different scenario concurrently, speeding up the process and reducing the time for decision between the alternative scenarios proposed by the agents. This approach was tested with a farmer agent that seeks optimal combinations of bivalve seeding areas in a large mariculture region, maximizing the production without exceeding the total allowed seeding area. Results obtained showed that the time needed to acquire a "near" optimal solution decreases proportionally with the number of simulators in the network, improving the performance of the agent's optimization process, without compromising its rationality. This work is a step forward towards an agent based decision support system to optimize complex environmental problems.

Abstract
Simulation is a very useful tool to gather new information about an implemented model, because it can run artificial environments instead of putting in risk some entities that are influenced in the real process. The simulation of physical, chemical and biological processes in coastal ecosystems is used as a way to understand the system internal dynamics and to predict its evolution over time, in order to promote behaviors environmentally friendly and to induce effective and efficient management of the ecosystem as a whole. However, there are several ways of translating and interpreting the data provided by the simulation such as applying appropriate data mining models. This paper describes an approach that uses a Decision Tree model to produce intuitive information about the influence of several environmental variables on the growth conditions of bivalve species within an aquaculture exploration in a coastal ecosystem. This information is captured by relating the values of simulated variables, like water temperature or organic matter, with the length of the bivalve's shell, extrapolating information about the organic or physical conditions that increase or decrease the growth of the bivalve species, and guiding the stakeholders to locations for the best practice of the seeding process.

Abstract
This document introduces ECOLANG, a communication language used for simulations of complex ecological systems. The language was developed with the main purpose of interchanging information between a simulation application of aquatic ecosystems (EcoDynamo) and several external agents. These agents have some goals about the simulation results or the simulated system. Examples of the former may be a calibration agent, with the goal of optimising the fit between observed and simulated results, whereas an example of the latter may be an aquaculture/farming agent, looking for production optimisation. This document focus on messages exchanged between EcoDynamo and the two mentioned agents providing examples of the communication protocol. This work is part of a larger project, where ECOLANG will be used as a tool for "goal-oriented intelligent simulations", towards sustainable management of coastal ecosystems.

Abstract
Software packages for educational purposes are increasingly important for classroom use, because they are a valuable way to motivate learning. However, the resources available on the market, with the designation of educational software, often have dubious quality. This led the authors to reflect on the process of evaluating the quality of a given educational software and develop tools that allow to measure its quality. This paper presents the results of the technical (software) and educational (guidelines for the didactic exploration) evaluation of Courseware Sere - The Human Being and Natural Resources - made by students of 2nd cycle of basic education (10 to 12 year old). The results indicate that the resource is easy to use, even for students who do not regularly use computers. It can also be concluded that the evaluation of educational resources, made by end-users, allows to correct issues related to usability (navigation and interaction), and to ensure the adequacy of resource to the target levels of education.

Abstract
A structured vocabulary of terms, such as a textual thesaurus, provides a way to conceptually describe visual information. The retrieval model described in this paper combines a conceptual and a visual layer as a first step towards the integration of ontologies and content-based image retrieval. Terms are related with image regions through a weighted association. This model allows the execution of concept-level queries, fulfilling user expectations and reducing the so-called semantic gap. Region-based relevance feedback is used to improve the quality of results in each query session and to help in the discovery of associations between text and image. The learning mechanism, whose function is to discover existing term-region associations, is based on a clustering algorithm applied over the features space and on propagation functions, which acts in each cluster where new information is available from user interaction. This approach is validated with the presentation of promising results obtained using the VOIR - Visual Object Information Retrieval system.