Large ships are efficient in transporting oil and its derivatives. However, they can cause spills in the event of accidents. The aim of the study is to simulate oil dispersion processes in scenarios of likely accidents with ships traveling on sea routes interconnected with Amazonian ports and with the Atlantic Ocean. Navigation routes were based on traffic data to identify risk areas, as well as to compare them to data from the environmental (oil) sensitivity index and to results of numerical simulations of plumes dispersion. These three approaches were integrated to each other in order to assess potential environmental impacts of plumes on coastal biota and human populations. Scenarios results have indicated that the rainy season is the most critical period for plumes dispersion. But, depending on the emission point, plumes tend to remain close to the coast, extend up to 132 km within 72 h, affecting the biodiversity, protected areas and water supply systems from the urban areas.

Animal-Vehicle Collision (AVC) is a predominant problem in both urban and rural roads and highways. Detecting animals on the road is challenging due to factors like the fast movement of both animals and vehicles, highly cluttered environmental settings, noisy images, and occluded animals. Deep learning has been widely used for animal applications. However, they require large training data; henceforth, the dimensionality increases, leading to a complex model. In this paper, we present an animal detection system for mitigating AVC. The proposed system integrates sparse representation and deep features optimized with FixResNeXt. The deep features extracted from candidate parts of the animals are represented in a sparse form using a feature-efficient learning algorithm called Sparse Network of Winnows (SNoW). The experimental results prove that the proposed system is invariant to the viewpoint, partial occlusion, and illumination. On the benchmark datasets, the proposed system has achieved an average accuracy of 98.5%.

Air quality in urban areas attracts great attention due to increasing pollutant emissions and their negative effects on human health and environment. Numerous studies, such as those by Mouilleau and Champassith (J Loss Prevent Proc 22(3): 316–323, 2009), Xie et al. (J Hydrodyn 21(1): 108–117, 2009), and Yassin (Environ Sci Pollut Res 20(6): 3975–3988, 2013) focus on the air pollutant dispersion with no buoyancy effect or weak buoyancy effect. A few studies, such as those by Hu et al. (J Hazard Mater 166(1): 394–406, 2009; J Hazard Mater 192(3): 940–948, 2011; J Civ Eng Manag (2013)) focus on the fire-induced dispersion of pollutants with heat buoyancy release rate in the range from 0.5 to 20 MW. However, the air pollution source might very often be concentrated and intensive, as a consequence of the hazardous materials fire. Namely, transportation of fuel through urban areas occurs regularly, because it is often impossible to find alternative supply routes. It is accompanied with the risk of fire accident occurrences. Accident prevention strategies require analysis of the worst scenarios in which fire products jeopardize the exposed population and environment. The aim of this article is to analyze the impact of wind flow on air pollution and human vulnerability to fire products in a street canyon. For simulation of the gasoline tanker truck fire as a result of a multivehicle accident, computational fluid dynamics large eddy simulation method has been used. Numerical results show that the fire products flow vertically upward, without touching the walls of the buildings in the absence of wind. However, when the wind velocity reaches the critical value, the products touch the walls of the buildings on both sides of the street canyon. The concentrations of carbon monoxide and soot decrease, whereas carbon dioxide concentration increases with the rise of height above the street canyon ground level. The longitudinal concentration of the pollutants inside the street increases with the rise of the wind velocity at the roof level of the street canyon.