基于长短时记忆网络的水质预测模型研究

在当前水质数据急剧增加的背景下,为了挖掘水质时间序列中的更多信息,提升水质预测的精度,构建了基于缺失值填补算法和长短时记忆网络(LSTM)相结合的水质预测模型。通过缺失值填补算法进行水质数据的缺失值处理,利用LSTM网络分别构建不同水质参数的预测模型,以太湖水质监测数据为样本,对模型进行检验。结果表明,基于缺失值填补算法-LSTM的水质预测模型适应性强,相较传统SVM、BP神经网络、RNN、LSTM模型预测精度更高,对水环境保护具有重要意义。     

Characterizing fish responses to a river restoration over 21 years based on species’ traits

Understanding restoration effectiveness is often impaired by a lack of high‐quality, long‐term monitoring data and, to date, few researchers have used species’ trait information to gain insight into the processes that drive the reaction of fish communities to restoration. We examined fish‐community responses with a highly resolved data set from 21 consecutive years of electrofishing (4 years prerestoration and 17 years postrestoration) at multiple restored and unrestored reaches from a river restoration project on the Lippe River, Germany. Fish abundance peaked in the third year after the restoration; abundance was 6 times higher than before the restoration. After 5–7 years, species richness and abundance stabilized at 2 and 3.5 times higher levels relative to the prerestoration level, respectively. However, interannual variability of species richness and abundance remained considerable, illustrating the challenge of reliably assessing restoration outcomes based on data from individual samplings, especially in the first years following restoration. Life‐history and reproduction‐related traits best explained differences in species’ responses to restoration. Opportunistic short‐lived species with early female maturity and multiple spawning runs per year exhibited the strongest increase in abundance, which reflected their ability to rapidly colonize new habitats. These often small‐bodied and fusiform fishes typically live in dynamic and ephemeral instream and floodplain areas that river‐habitat restorations often aim to create, and in this case their increases in abundance indicated successful restoration. Our results suggest that a greater consideration of species’ traits may enhance the causal understanding of community processes and the coupling of restoration to functional ecology. Trait‐based assessments of restoration outcomes would furthermore allow for easier transfer of knowledge across biogeographic borders than studies based on taxonomy.     

Identifying impediments to long‐distance mammal migrations

In much of the world, the persistence of long‐distance migrations by mammals is threatened by development. Even where human population density is relatively low, there are roads, fencing, and energy development that present barriers to animal movement. If we are to conserve species that rely on long‐distance migration, then it is critical that we identify existing migration impediments. To delineate stopover sites associated with anthropogenic development, we applied Brownian bridge movement models to high‐frequency locations of pronghorn (Antilocapra americana) in the Greater Yellowstone Ecosystem. We then used resource utilization functions to assess the threats to long‐distance migration of pronghorn that were due to fences and highways. Migrating pronghorn avoided dense developments of natural gas fields. Highways with relatively high volumes of traffic and woven‐wire sheep fence acted as complete barriers. At crossings with known migration bottlenecks, use of high?quality forage and shrub habitat by pronghorn as they approached the highway was lower than expected based on availability of those resources. In contrast, pronghorn consistently utilized high?quality forage close to the highway at crossings with no known migration bottlenecks. Our findings demonstrate the importance of minimizing development in migration corridors in the future and of mitigating existing pressure on migratory animals by removing barriers, reducing the development footprint, or installing crossing structures. Identificación de los Impedimentos para las Migraciones de Larga Distancia de Mamíferos     

Declines in migrant shorebird populations from a winter‐quarter perspective

Many long‐distance migrating shorebird (i.e., sandpipers, plovers, flamingos, oystercatchers) populations are declining. Although regular shorebird monitoring programs exist worldwide, most estimates of shorebird population trends and sizes are poor or nonexistent. We built a state‐space model to estimate shorebird population trends. Compared with more commonly used methods of trend estimation, state‐space models are more mechanistic, allow for the separation of observation and state process, and can easily accommodate multivariate time series and nonlinear trends. We fitted the model to count data collected from 1990 to 2013 on 18 common shorebirds at the 2 largest coastal wetlands in southern Africa, Sandwich Harbour (a relatively pristine bay) and Walvis Bay (an international harbor), Namibia. Four of the 12 long‐distance migrant species declined since 1990: Ruddy Turnstone (Arenaria interpres), Little Stint (Calidris minuta), Common Ringed Plover (Charadrius hiaticula), and Red Knot (Calidris canutus). Populations of resident species and short‐distance migrants increased or were stable. Similar patterns at a key South African wetland suggest that shorebird populations migrating to southern Africa are declining in line with the global decline, but local conditions in southern Africa's largest wetlands are not contributing to these declines. State‐space models provide estimates of population levels and trends and could be used widely to improve the current state of water bird estimates.     

面向密闭空间内外温度的时序预测模型

目的 研究密闭空间条件下已知外部温度时间序列对内部实时温度的预测推理问题。方法 选取密闭空间内外温度时序预测典型场景,抽象为多变量时间序列预测问题,分析变量间的关联性和依赖性。借鉴特征融合、注意力机制、多任务模型等思路,结合物理机制与数据特征,基于长短期记忆网络基本网络单元,构建密闭空间内外温度时序预测模型,并在万宁、敦煌、漠河对某型密闭空间进行数据采样,基于三地数据集进行不同模型试验。结果 多变量模型比单变量模型具有更好性能,注意力机制对该场景没有显著性能提升,结合物理机制的模型结构设计充分考虑了变量之间的关联性和依赖性,能显著提升预测精度,双输入双输出的多变量时序预测模型具有相对最高的精度和最稳定的鲁棒性,是面向密闭空间内外温度时序预测的相对最优模型。结论 研究结论可指导密闭空间其他环境特征建模,研究思路可为其他多变量时序建模问题中变量之间的关联性、依赖性分析提供参考。     

PACIFIC NORTHWEST REGIONAL ASSESSMENT: THE IMPACTS OF CLIMATE VARIABILITY AND CLIMATE CHANGE ON THE WATER RESOURCES OF THE COLUMBIA RIVER BASIN1

ABSTRACT: The Pacific Northwest (PNW) regional assessment is an integrated examination of the consequences of natural climate variability and projected future climate change for the natural and human systems of the region. The assessment currently focuses on four sectors: hydrology/water resources, forests and forestry, aquatic ecosystems, and coastal activities. The assessment begins by identifying and elucidating the natural patterns of climate vanability in the PNW on interannual to decadal timescales. The pathways through which these climate variations are manifested and the resultant impacts on the natural and human systems of the region are investigated. Knowledge of these pathways allows an analysis of the potential impacts of future climate change, as defined by IPCC climate change scenarios. In this paper, we examine the sensitivity, adaptability and vulnerability of hydrology and water resources to climate variability and change. We focus on the Columbia River Basin, which covers approximately 75 percent of the PNW and is the basis for the dominant water resources system of the PNW. The water resources system of the Columbia River is sensitive to climate variability, especially with respect to drought. Management inertia and the lack of a centralized authority coordinating all uses of the resource impede adaptability to drought and optimization of water distribution. Climate change projections suggest exacerbated conditions of conflict between users as a result of low summertime streamfiow conditions. An understanding of the patterns and consequences of regional climate variability is crucial to developing an adequate response to future changes in climate.     

淮阳龙湖湿地保护及其资源开发的问题和对策研究

湿地被喻为“地球之肾”,因其蕴藏着极其丰富的生物资源,同时又具有汇聚、平衡水分、分配二氧化碳的功能.为充分发挥龙湖湿地的资源优势和旅游潜力,实现区域经济和生态环境保护的双赢,通过实地抽样调查、数据归类分析等方法,初步探讨了龙湖湿地的资源优势和面临的生态环境问题.建议从调控城市湿地生态系统与周边陆域生态系统的耦合关系入手,运用强化湿地管理、加大科技投入、提高群众保护意识等手段,构建一个和谐的生态环境.     

长距离输煤管道储浆罐与石油储罐对比研究

对比研究长距离输煤管道储浆罐与石油储罐搅拌设计特点、罐顶设计特点、消防安全设施特点。经分析可知,长距离输煤管道储浆罐宜选用推进式搅拌器,罐壁上设置导流结构,宜采用固定顶、锥顶,设置简单消防安全设施,宜选用难燃保温材料;石油储罐优先选用旋转喷射式搅拌器,罐内油品进行定期搅拌,采用浮顶,大型的一般选用外浮顶,需设置完善消防安全设施。     

Occupational safety conditions of bus drivers in Metro Manila,the Philippines

The study looks into the occupational safety and working conditions among bus drivers in Metro Manila, the Philippines. Quantitative data were collected through survey interviews of 95 bus drivers using the stratified sampling technique. Results showed that bus drivers worked an average of 16?h/day and were engaged in risky driving behaviors such as over-speeding and road racing in order to reach their quota for the day. Fifty-nine percent experienced work-related accidents, with a mean of three accidents. The most common accident was hitting another vehicle followed by side swipe. The accidents were blamed on other drivers, followed by vehicle defect, inattentiveness and tiredness/micro-sleep or sudden involuntary sleep while driving. The most common health symptoms experienced by the bus drivers were fatigue, back pain, and cough and colds. This study underlines the need for an occupational health and safety program for bus drivers in the Philippines.     

扎龙湿地表层沉积物多环芳烃的污染特征研究

利用索氏抽提、硅胶氧化铝净化和GC-MS定性和定量的方法研究了黑龙江省扎龙湿地12个表层沉积物中15种优控PAHs的环境行为,目的是阐明扎龙湿地表层沉积物中PAHs的污染水平、组成特征、污染来源和途径,评价PAHs对扎龙湿地水生生物的生态风险.结果表明,15种PAHs的总含量（干重）范围在31.9～290 ng/g之间...