Search

The characterization of the content and source of heavy metals are essential to assess the potential threat of metals to human health. The present study collected 140 topsoil samples around a Cu-Mo mine (Wunugetushan, China) and investigated the concentrations and spatial distribution pattern of Cr, Ni, Zn, Cu, Mo and Cd in soil using multivariate and geostatistical analytical methods. Results indicated that the average concentrations of six heavy metals, especially Cu and Mo, were obviously higher than the local background values. Correlation analysis and principal component analysis divided these metals into three groups, including Cr and Ni, Cu and Mo, Zn and Cd. Meanwhile, the spatial distribution maps of heavy metals indicated that Cr and Ni in soil were no notable anthropogenic inputs and mainly controlled by natural factors because their spatial maps exhibited non-point source contamination. The concentrations of Cu and Mo gradually decreased with distance away from the mine area, suggesting that human mining activities may be crucial in the spreading of contaminants. Soil contamination of Zn were associated with livestock manure produced from grazing. In addition, the environmental risk of heavy metal pollution was assessed by geo-accumulation index. All the results revealed that the spatial distribution of heavy metals in soil were in agreement with the local human activities. Investigating and identifying the origin of heavy metals in pasture soil will lay the foundation for taking effective measures to preserve soil from the long-term accumulation of heavy metals., (Copyright © 2018 Elsevier Inc. All rights reserved.)

Glaciers, particularly alpine glaciers, have been receding globally at an accelerated rate in recent decades. The glacial melt-induced release of pollutants (e.g., mercury) and its potential impact on the atmosphere and glacier-fed ecosystems has drawn increasing concerns. During 15th-20th August, 2011, an intensive sampling campaign was conducted in Qugaqie Basin (QB), a typical high mountain glacierized catchment in the inland Tibetan Plateau, to investigate the export and transport of mercury from glacier to runoff. The total mercury (THg) level in Zhadang (ZD) glacier ranged from <1 to 20.8 ng L -1 , and was slightly higher than levels measured in glacier melt water and the glacier-fed river. Particulate Hg (PHg) was the predominant form of Hg in all sampled environmental matrices. Mercury concentration in Qugaqie River (QR) was characterized by a clear diurnal variation which is linked to glacier melt. The estimated annual Hg exports by ZD glacier, the upper river basin and the entire QB were 8.76, 7.3 and 157.85 g, respectively, with respective yields of 4.61, 0.99 and 2.74 μg m -2  yr -1 . Unique landforms and significant gradients from the glacier terminus to QB estuary might promote weathering and erosion, thereby controlling the transport of total suspended particulates (TSP) and PHg. In comparison with other glacier-fed rivers, QB has a small Hg export yet remarkably high Hg yield, underlining the significant impact of melting alpine glaciers on regional Hg biogeochemical cycles. Such impacts are expected to be enhanced in high altitude regions under the changing climate., (Copyright © 2016 Elsevier Ltd. All rights reserved.)

To investigate the transport of mercury from glacier to runoff in typical inland glacial area in the Tibetan Plateau, we selected Zhadang glacier and Qugaqie river Basin located in the Nyainqentanglha Range region and collected samples from snow pit, glacier melt-water and Qugaqie river water during 15th August to 9'h September 2011. Mercury speciation and concentrations were determined and their distribution and controlling factors in different environmental compartments were analyzed. The results showed that the average THg concentrations were (3.79 +/- 5.12) ng x L(-1), (1.06 +/- 0.77) ng x L(-1) and (1.02 +/- 0.24) ng x L(-1) for glacier snow, glacier melt-water and Qugaqie river water, respectively, all of which were at the global background levels. Particulate-bound mercury accounted for large proportion of mercury in all environmental matrices, while mercury in glacial melt-water was controlled by total suspended particle, and mercury in Qugaqie river water co-varied with runoff. With the increase of temperature, glacier melted and released water as well as mercury into glacier-fed river. Total mercury concentrations in glacier melt water, upstream and downstream peaked at 14:00, 16:00 and after 20:00, respectively, reflecting the process of mercury release from glacier and its subsequent transport in the glacier fed river. The transport of riverine mercury was controlled by multiple factors. Under the context of climate change, glacier ablation and the increasing runoff will play increasingly important roles in mercury release and transport.

Mercury (Hg) in aquatic ecosystems is of great concern due to its toxicity, bioaccumulation, and magnification in the food web. The Tibetan Plateau (TP) is endowed with the highest and largest lakes on earth, whereas Hg distribution and behavior in lake waters are least known. In this study, surface water samples from 38 lakes over the TP were collected and determined for the total Hg (THg) concentrations. Results revealed a wide range of THg concentrations from <1 ng to 40.3 ng L(-1). THg in lake waters exhibited an increasing trend along the southeast to northwest transect over the TP. Strong positive correlations were observed between THg concentrations and salinity and salinity-related environmental variables, especially for total dissolved solids (TDS) and some of the major ions such as Na(+), K(+), and Cl(-), suggesting the enrichment of Hg in saline lakes. The large-scale geographical pattern of climatic and environmental factors shows a decreasing precipitation and an increasing evaporation northwards and westwards and thereby induces gradient-enhanced enrichment of soluble substances in lake waters, which are likely to complex more Hg in northwestern TP. Our study provides the first comprehensive baseline data set of Hg in Tibetan lake waters and highlights the concurrent high Hg and salinity, representing valuable references and fundamental rules in further understanding the behavior and fate of Hg in lakes over the TP and perhaps high-altitude regions beyond.

In order to investigate the temporal and spatial variations, sources, and major controlling factors of the major ions in Nam Co Lake water, inshore surface water samples were collected at a fixed site (30 degrees 47.27'N, 90 degrees 58.53'E, 4718 m a. s. l.) from 2006 to 2010, at the vertical profiles in the center of the lake in August 2009, and at both the vertical profiles in the center of the lake and at the surface layers of different sites in the Nam Co Lake in October 2010. The results indicated that Na+ was the dominant cation and HCO3- was the dominant anion in the lake water. The concentrations of most ions were higher in monsoon seasons (June - September) and lower in non-monsoon seasons, especially when the lake was frozen (January -April). However, the Ca2+ concentration showed a reverse trend of seasonal variations, namely, higher values in the frozen period and lower in monsoon seasons. Analysis of water samples collected from the vertical profiles indicated that the concentrations of all ions except Ca2+ increased with the depth in nonmonsoon seasons (e. g. October). The major ions in Nam Co Lake were mainly contributed by river input. There were a variety of factors that influenced the temporal and spatial variations of the major ions in the Nam Co Lake, such as evaporation, precipitation, pH values, etc., among which, evaporation was the most important controlling factor, causing the increasing Na+ concentration and decreasing Ca2+ concentration in the lake water.

To investigate the tempo-spatial distribution of total mercury (T-Hg) concentration in water bodies in the Nam Co basin on the Tibetan Plateau, inflowing river water and surface lake water samples were collected from 2007 to 2010. The T-Hg concentration and its relationship with precipitation and river runoff were analyzed. The results showed that the average T-Hg concentration was (1.09 +/- 0.73) ng x L(-1) and (2.87 +/- 2.59) ng x L(-1) for surface lake water and river water, respectively, both of which were significantly lower than those of Hg contaminated waters. T-Hg concentration in off-shore lake water was much higher during the monsoon season than in the non-monsoon season, and its level and spatial variation were significantly greater than those in central lake water. T-Hg concentration in river water showed significant seasonal variations with the highest values during the monsoon season and the lowest during the post-monsoon season, which were in accordance with the variations of precipitation. A fixed point observation at Niyaqu River indicated that the temporal changes of the T-Hg concentrations in river water were in accordance with those of the runoff. The spatial distribution features of T-Hg concentrations in inflowing river water varied in different periods, possibly resulting from the differences in drainage areas, background mercury levels in soils, and water supplies for rivers at different locations of the Nam Co basin.

Simple Summary: Identifying bird species is crucial in various bird monitoring tasks. In this study, we implemented a bird species recognition method based on visual features by convolutional neural networks. This paper proposed an attention-guided data enhancement method and a decoupled knowledge distillation model compression method. Using the two methods, we created a novel and efficient lightweight fine-grained bird classification model. The model not only achieved high accuracy in bird classification but was also user-friendly for edge devices such as mobile. Our work may be useful for bird species recognition and bird biodiversity monitoring. Classifying birds accurately is essential for ecological monitoring. In recent years, bird image classification has become an emerging method for bird recognition. However, the bird image classification task needs to face the challenges of high intraclass variance and low inter-class variance among birds, as well as low model efficiency. In this paper, we propose a fine-grained bird classification method based on attention and decoupled knowledge distillation. First of all, we propose an attention-guided data augmentation method. Specifically, the method obtains images of the object's key part regions through attention. It enables the model to learn and distinguish fine features. At the same time, based on the localization–recognition method, the bird category is predicted using the object image with finer features, which reduces the influence of background noise. In addition, we propose a model compression method of decoupled knowledge distillation. We distill the target and nontarget class knowledge separately to eliminate the influence of the target class prediction results on the transfer of the nontarget class knowledge. This approach achieves efficient model compression. With 67% fewer parameters and only 1.2 G of computation, the model proposed in this paper still has a 87.6% success rate, while improving the model inference speed. [ABSTRACT FROM AUTHOR]