Search

• A comprehensive assessment of the three major droughts' impacts on Amazon forests. • The 2015/16 drought is the most severe meteorological drought since 2000. • Drought in 2015/16 showed lagging impacts on photosynthesis, greenness, and leaf area. • Photosynthesis, greenness, and leaf area had similar suppression in 2010 and 2015/16. • Canopy water content exhibited more widespread and severe drought impacts in 2015/16. The Amazon forests play an integral role in the global carbon cycle and have a substantial impact on Earth's climate. However, it is increasingly susceptible to the effects of prolonged droughts, exacerbated by climate change and human activities. This vulnerability underscores the importance of understanding the forests' reaction to such environmental stressors. Despite their significance, comprehensive cross-comparisons of the climate and vegetation responses during the 2005, 2010, and 2015/2016 drought episodes are not well-established. Here we utilize a range of gridded vegetation and climate datasets—including leaf area index (LAI), solar-induced chlorophyll fluorescence (SIF), enhanced vegetation index (EVI), vegetation optical depth (VOD), self-calibrating Palmer drought severity index (scPDSI), precipitation (P), land surface temperature (LST), and photosynthetically active radiation (PAR)—to thoroughly assess the climate and vegetation response to these three drought events. Our findings reveal that the extent of drought inhibition in the Amazon forests was 74.7 % in 2015, increasing to 81.3 % in 2016, a significant escalation from 49.6 % in 2005 and 57.7 % in 2010. The effects of these three droughts on vegetation varied in both physiological and structural aspects. The Amazon forests' photosynthetic activity, greenness, and leaf area experienced comparable suppression in 2010 and 2015/2016 droughts. However, canopy water content exhibited more extensive and severe impacts during the 2015/2016 drought. Our findings indicate that varying sensitivities to water deficit and solar radiation lead to diverse spatial patterns and intensities of vegetation response, highlighting the complex dynamics of the Amazon forests under drought stress. [ABSTRACT FROM AUTHOR]

In recent years, there has been a growing prevalence of deep learning in various domains, owing to advancements in information technology and computing power. Graph neural network methods within deep learning have shown remarkable capabilities in processing graph-structured data, such as social networks and traffic networks. As a result, they have garnered significant attention from researchers.However, real-world data often face challenges like data sparsity and missing labels, which can hinder the performance and generalization ability of graph convolutional neural networks. To overcome these challenges, our research aims to effectively extract the hidden features and topological information of graph convolutional neural networks. We propose an innovative model called Adaptive Feature and Topology Graph Convolutional Neural Network (AAGCN). By incorporating an adaptive layer, our model preprocesses the data and integrates the hidden features and topological information with the original data’s features and structure. These fused features are then utilized in the convolutional layer for training, significantly enhancing the expressive power of graph convolutional neural networks.To evaluate the effectiveness of the adaptive layer in the AAGCN model, we conducted node classification experiments on real datasets. The results validate its ability to address data sparsity and improve the classification performance of graph convolutional neural networks.In conclusion, our research primarily focuses on addressing data sparsity and missing labels in graph convolutional neural networks. The proposed AAGCN model, which incorporates an adaptive layer, effectively extracts hidden features and topological information, thereby enhancing the expressive power and classification performance of these networks. [ABSTRACT FROM AUTHOR]