AAGCN: a graph convolutional neural network with adaptive feature and topology learning.

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.

China's current forest age structure will lead to weakened carbon sinks in the near future.

Forests are chiefly responsible for the terrestrial carbon sink that greatly reduces the buildup of CO2 concentrations in the atmosphere and alleviates climate change. Current predictions of terrestrial carbon sinks in the future have so far ignored the variation of forest carbon uptake with forest age. Here, we predict the role of China's current forest age in future carbon sink capacity by generating a high-resolution (30 m) forest age map in 2019 over China's landmass using satellite and forest inventory data and deriving forest growth curves using measurements of forest biomass and age in 3,121 plots. As China's forests currently have large proportions of young and middle-age stands, we project that China's forests will maintain high growth rates for about 15 years. However, as the forests grow older, their net primary productivity will decline by 5.0% ± 1.4% in 2050, 8.4% ± 1.6% in 2060, and 16.6% ± 2.8% in 2100, indicating weakened carbon sinks in the near future. The weakening of forest carbon sinks can be potentially mitigated by optimizing forest age structure through selective logging and implementing new or improved afforestation. This finding is important not only for the global carbon cycle and climate projections but also for developing forest management strategies to enhance land sinks by alleviating the age effect.

Drylands contribute disproportionately to observed global productivity increases.

Drylands cover about 40% of the terrestrial surface and are sensitive to climate change, but their relative contributions to global vegetation greening and productivity increase in recent decades are still poorly known. Here, by integrating satellite data and biosphere modeling, we showed that drylands contributed more to global gross primary productivity (GPP) increase (65% ± 16%) than to Earth greening (33% ± 15%) observed during 1982-2015. The enhanced productivity per unit leaf area, i.e., light-use efficiency (LUE), was the mechanism behind this pattern. We also found that LUE was more sensitive to soil moisture than to atmospheric vapor pressure deficit (VPD) in drylands, while the opposite was observed (i.e., LUE was more sensitive to VPD) in humid areas. Our findings suggest the importance of using different moisture stress metrics in projecting the vegetation productivity changes of dry versus humid regions and highlight the prominent role of drylands as key controllers of the global carbon cycle.

Stable isotope variations of dew under three different climates.

As a supplementary or the only water source in dry regions, dew plays a critical role in the survival of organisms. The new hydrological tracer 17O-excess, with almost sole dependence on relative humidity, provides a new way to distinguish the evaporation processes and reconstruct the paleoclimate. Up to now, there is no published daily dew isotope record on δ2H, δ18O, δ17O, d-excess, and 17O-excess. Here, we collected daily dew between July 2014 and April 2018 from three distinct climatic regions (i.e., Gobabeb in the central Namib Desert with desert climate, Nice in France with Mediterranean climate, and Indianapolis in the central United States with humid continental climate). The δ2H, δ18O, and δ17O of dew were simultaneously analyzed using a Triple Water Vapor Isotope Analyzer based on Off-Axis Integrated Cavity Output Spectroscopy technique, and then d-excess and 17O-excess were calculated. This report presents daily dew isotope dataset under three climatic regions. It is useful for researchers to use it as a reference when studying global dew dynamics and dew formation mechanisms.

Observed increasing water constraint on vegetation growth over the last three decades.

Despite the growing interest in predicting global and regional trends in vegetation productivity in response to a changing climate, changes in water constraint on vegetation productivity (i.e., water limitations on vegetation growth) remain poorly understood. Here we conduct a comprehensive evaluation of changes in water constraint on vegetation growth in the extratropical Northern Hemisphere between 1982 and 2015. We document a significant increase in vegetation water constraint over this period. Remarkably divergent trends were found with vegetation water deficit areas significantly expanding, and water surplus areas significantly shrinking. The increase in water constraints associated with water deficit was also consistent with a decreasing response time to water scarcity, suggesting a stronger susceptibility of vegetation to drought. We also observed shortened water surplus period for water surplus areas, suggesting a shortened exposure to water surplus associated with humid conditions. These observed changes were found to be attributable to trends in temperature, solar radiation, precipitation, and atmospheric CO2. Our findings highlight the need for a more explicit consideration of the influence of water constraints on regional and global vegetation under a warming climate.

A semi-analytical algorithm for deriving the particle size distribution slope of turbid inland water based on OLCI data: A case study in Lake Hongze.

The particle size distribution (PSD) slope (ξ) can indicate the predominant particle size, material composition, and inherent optical properties (IOPs) of inland waters. However, few semi-analytical methods have been proposed for deriving ξ from the surface remote sensing reflectance due to the variable optical state of inland waters. A semi-analytical algorithm was developed for inland waters having a wide range of turbidity and ξ in this study. Application of the proposed model to Ocean and Land Color Instrument (OLCI) imagery of the water body resulted in several important observations: (1) the proposed algorithm (754 nm and 779 nm combination) was capable of retrieving ξ with R2 being 0.72 (p < 0.01, n = 60), and MAPE and RMSE being 4.37% and 0.22 (n = 30) respectively; (2) the ξ in HZL was lower in summer than other seasons during the period considered, this variation was driven by the phenological cycle of algae and the runoff caused by rainfall; (3) the band optimization proposed in this study is important for calculating the particle backscattering slope (η) and deriving ξ because it is feasible for both algae dominant and sediment governed turbid inland lakes. These observations help improve our understanding of the relationship between IOPs and ξ, which are affected by different bio-optic processes and algal phenology in the lake environment.

Triple isotope variations of monthly tap water in China.

Tap water isotopic compositions could potentially record information on local climate and water management practices. A new water isotope tracer 17O-excess became available in recent years providing additional information of the various hydrological processes. Detailed data records of tap water 17O-excess have not been reported. In this report, monthly tap water samples (n = 652) were collected from December 2014 to November 2015 from 92 collection sites across China. The isotopic composition (δ2H, δ18O, and δ17O) of tap water was analyzed by a Triple Water Vapor Isotope Analyzer (T-WVIA) based on Off-Axis Integrated Cavity Output Spectroscopy (OA-ICOS) technique and two second-order isotopic variables (d-excess and 17O-excess) were calculated. The geographic location information of the 92 collection sites including latitude, longitude, and elevation were also provided in this dataset. This report presents national-scale tap water isotope dataset at monthly time scale. Researchers and water resource managers who focus on the tap water issues could use them to probe the water source and water management strategies at large spatial scales.