No abstract

Oil spill detection and mapping using deep learning (OSDMDL) is crucial for assessing its impact on coastal and marine ecosystems. A novel approach was employed in this study to evaluate the scientific literature in this field through bibliometric analysis and literature review. The Scopus database was used to evaluate the relevant scientific literature in this field, followed by a bibliometric analysis to extract additional information, such as architecture type, country collaboration, and most cited papers. The findings highlight significant advancements in oil detection at sea, with a strong correlation between technological evolution in detection methods and improved remote sensing data acquisition. Multilayer perceptrons (MLP) emerged as the most prominent neural network architecture in 11 studies, followed by a convolutional neural network (CNN) in 5 studies. U-Net, DeepLabv3+, and fully convolutional network (FCN) were each used in three studies, demonstrating their relative significance too. The analysis provides insights into collaboration, interdisciplinarity, and research methodology and contributes to the development of more effective policies, strategies, and technologies for mitigating the environmental impact of oil spills in OSDMDL.

Background Fire dynamics in the Amazon, while not fully understood, are central to designing fire management strategies and providing a baseline for projecting the effects of climate change. Aims The study investigates the recent fire probabilities in the northeastern Amazon and project future ‘fire niches’ under global warming scenarios, allowing the evaluation of drivers and areas of greatest susceptibility. Methods Using the maximum entropy method, we combined a complex set of predictors with fire occurrences detected during 2000–2020. We estimated changes in fire patterns in the near (2020–2040) and distant (2080–2100) future, under two contrasting scenarios of shared socioeconomic pathways. Key results Based on current conditions, the spatial fire pattern is affected by farming activities and fire is more common in savannas than in forests. Over long time scales, changes toward a warmer and drier climate, independent of land cover change, are expected to create conditions more conducive to burning. Conclusion and implications Our study helps in understanding the multiple ecological and human interactions that result in different fire regimes in the Amazon. Future efforts can improve outcomes through more complex models that couple predictions of land use and land cover changes, shifts in vegetation resulting from climate change and fires, and fuel dynamics.