Abstract
Climate change increases the average temperature and reduces precipitation, leading to an increased risk of rural fires around the world, but mainly in regions with a Mediterranean climate, such as Portugal. Despite the high risk of rural fires, fire is still a traditional land management practice. Beyond fire misused risk, the accumulation of fuel loads, due to the high population dispersion and lower interaction of communities with the land, also drives rural fires risks. Thus, researchers have sought to understand the key features of communication practices to achieve the most desired natural risk management results in relation to rural fires. According to the Committee on Risk Perception and Communication, in 1989, risk communication is defined as “an interactive process of exchanging information and opinions between individuals, groups, and institutions�. However, there is a need for dialogue between the responsible for communication and relevant stakeholders. The Carnegie Mellon mental-models approach encompasses participatory processes to translate the nature and magnitude of the risk, allowing for a deeper understanding of what can be done to mitigate social and environmental impacts in the future. Considering rural fires risks and risk communication challenges, this study aims to design an effective risk communication plan oriented to the rural population, in the context of rural fires. For this objective to be achieved, the mental models approach was developed, exploring the underlying reasons for resistance to behavioral change and defining guidelines to support the design of new risk communication strategies, including the dissemination of new behaviors and practices that mitigate the ignition of rural fires. This systemic approach has been applied and tested in the rePLANT project, whose purpose is the development of research activities, innovation and transfer of knowledge and technology, to increase sustainable forest management, the competitiveness of the Portuguese forestry sector and reduce the impact of rural fires. Preliminary findings show what must be communicated about rural fires risk, how must be communicated, target audiences' profiles, the role of actors for effective risk communication on rural fires, attributes of good risk communication on rural fires, evaluation of the risk communication results. Finally, it is expected that this research provides guidelines to help decision-makers and stakeholders to design an effective risk communication plan oriented to prevent traditional burning practices, as well as to mitigate socioeconomic and environmental impacts in the future. This approach also highlights the importance of exploring the underlying reasons for resistance to behavioral change and defining guidelines to support the design of new risk communication strategies.

Abstract
Social media are transforming relationships with customers for all sectors, including tourism. Since the search for information is a critical aspect of tourist purchase decision process, the importance of social media for tourism is evident. However, the presence of tourism brands in social media is not enough to have an impact on tourist purchase decisions: it is necessary to generate engagement. This chapter aims to conceptualize tourist engagement on social media and identify tourist engagement indicators. Tourist engagement was conceptualized through a literature review that identified four dimensions of engagement: popularity, commitment, virality, and post engagement. A set of indicators is proposed to measure tourist engagement in each of these dimensions. The proposed TSM engagement framework was validated through a mixed-method approach, using secondary data and interviews carried out with Brazilian tourist destinations.

Abstract
Guiding the building occupants under fire emergency to a safe place is an open research problem. Finding solutions to address the problem requires a perfect knowledge of the fire building evacuation domain. The use of ontologies to model knowledge of a domain allows a common and shared understanding of that domain, between people and heterogeneous systems. This paper presents an ontology that aims to build a knowledge model to better understand the referred domain and to help develop more capable building evacuation solutions and systems. The herein proposed ontology considers the different variables and actors involved in the fire building evacuation process. We followed the Methontology methodology for its developing, and we present all the development steps, from the specification to its implementation with the Protégé tool.

Abstract
Seeking to "ensure availability and sustainable management of water and sanitation for all" is an admirable Sustainable Development Goal and an honourable commitment of the United Nations and its Member States regarding the human right to safe drinking water and sanitation services (WSSs). However, the majority of countries are not on target to achieve this by 2030, with several of them moving away from the best practices. Brazil is one of these cases, given, for example, the existing asymmetries in the access to water supply and sanitation service networks. For this reason, we propose a benchmarking exercise using a two-stage Data Envelopment Analysis to measure the technical and scale efficiency of the Brazilian municipalities' WSSs, noting their contextual environment. Our results point towards low mean efficiency scores, motivated by the existence of significant scale inefficiencies (the vast majority of municipalities are operating at a larger than optimal scale). Furthermore, the Water source was found to be a statistically significant efficiency predictor, with statistically significant differences found in terms of Ownership and Geography. Ultimately, we suggest policy-making and regulatory possibilities based on debureaucratization, the implementation of stricter expenditure control policies, and investment in the expansion of WSSs.

Abstract
Morphing attacks are one of the many threats that are constantly affecting deep face recognition systems. It consists of selecting two faces from different individuals and fusing them into a final image that contains the identity information of both. In this work, we propose a novel regularisation term that takes into account the existent identity information in both and promotes the creation of two orthogonal latent vectors. We evaluate our proposed method (OrthoMAD) in five different types of morphing in the FRLL dataset and evaluate the performance of our model when trained on five distinct datasets. With a small ResNet-18 as the backbone, we achieve state-of-the-art results in the majority of the experiments, and competitive results in the others.

Abstract
The forecast of wind speed is prerequisite for wind power prediction, which is one of the most effective means of promoting wind power absorption. However, when modeling for wind speed sequences with different fluctuations, most existing researches ignore the influence of time scale of wind speed fluctuation period, let alone the low compatibility between training and testing samples that severely limit the training performance of forecasting model. To improve the accuracy of wind speed and wind power forecasting, an ultra-short-term wind speed forecasting model based on time scale recognition and dynamic adaptive modeling is proposed in this paper. First, a series of wind processes are divided from the historical wind speed sequence according to the natural variation characteristics of wind speed. Second, we divide all the wind processes into two patterns based on their time scale, and an SVC model with input features extracted from meteorological data is built to identify the time scale of the current wind process. Third, for a specifically identified wind process, the complex network algorithm is applied in data screening to select high compatible training samples to train the forecast model dynamically for current input. Simulation indicates that the proposed approach presents higher accuracy than benchmark models using the same forecasting algorithms but without considering the time scale and data screening.