The future monitoring capabilities of four network configurations in eastern Canada

The capability of deposition networks in eastern Canada to resolve regional trends has been examined by Blanchard et al. (1996, Atmospheric Environment 30, 2539–2549). This paper extends the earlier methods to permit evaluation of the need for specific individual sites, comparison of additional network configurations, and consideration of more regions. Parameter estimates for the statistical model are improved by using monthly instead of annual data. For the regions that we examined, eliminating all provincial sites in eastern Canada would cause an increase of 1 to 6 years in the time required for reaching a 90% probabiility of detecting the expected future trends. The elimination of certain key provincial sites may increase the uncertainties in the determination of deposition isopleths of particular interest, such as the 20 kg ha⁻¹ yr⁻¹ contours. The statistical techniques presented here are quite general and can be extended to statistical tests or estimators other than those illustrated here.     

Voluntary instruments for ambitious objectives? The experience of the EU Covenant of Mayors

ABSTRACT

Over the last decade, innovative governance architectures have been progressively promoted across European Union (EU) environmental and climate policies with the purpose of improving the effectiveness of intervention through better cross-sectoral policy integration and increased involvement of sub-state and non-governmental actors in the policy process. By combining the theoretical insights of polycentric governance and the concept of usage, the case of the Covenant of Mayors (CoM) is analysed to uncover the extent to which this voluntary programme has empowered local authorities within the EU strategy for sustainable energy by encouraging coordination and learning. This illustrates how a range of policy variables determined the dynamics of the programme’s implementation in Italy and the United Kingdom.     

MODELKEY. Models for assessing and forecasting the impact of environmental key pollutants on freshwater and marine ecosystems and biodiversity (5 pp)

BACKGROUND: Triggered by the requirement of Water Framework Directive for a good ecological status for European river systems till 2015 and by still existing lacks in tools for cause identification of insufficient ecological status MODELKEY (http:// www.modelkey.org), an Integrated Project with 26 partners from 14 European countries, was started in 2005. MODELKEY is the acronym for 'Models for assessing and forecasting the impact of environmental key pollutants on freshwater and marine ecosystems and biodiversity'. The project is funded by the European Commission within the Sixth Framework Programme. OBJECTIVES: MODELKEY comprises a multidisciplinary approach aiming at developing interlinked tools for an enhanced understanding of cause-effect-relationships between insufficient ecological status and environmental pollution as causative factor and for the assessment and forecasting of the risks of key pollutants on fresh water and marine ecosystems at a river basin and adjacent marine environment scale. New modelling tools for risk assessment including generic exposure assessment models, mechanistic models of toxic effects in simplified food chains, integrated diagnostic effect models based on community patterns, predictive component effect models applying artificial neural networks and GIS-based analysis of integrated risk indexes will be developed and linked to a user-friendly decision support system for the prioritisation of risks, contamination sources and contaminated sites. APPROACH: Modelling will be closely interlinked with extensive laboratory and field investigations. Early warning strategies on the basis of sub-lethal effects in vitro and in vivo are provided and combined with fractionation and analytical tools for effect-directed analysis of key toxicants. Integrated assessment of exposure and effects on biofilms, invertebrate and fish communities linking chemical analysis in water, sediment and biota with in vitro, in vivo and community level effect analysis is designed to provide data and conceptual understanding for risk arising from key toxicants in aquatic ecosystems and will be used for verification of various modelling approaches. CONCLUSION AND PERSPECTIVE: The developed tools will be verified in case studies representing European key areas including Mediterranean, Western and Central European river basins. An end-user-directed decision support system will be provided for cost-effective tool selection and appropriate risk and site prioritisation.     

关于职称评聘制度改革的几点思考

本文论述了现行职称评聘制度存在的主要问题。指出了深化职称评聘制度改革的必要 性,提出了深化职称评聘制度改革的几点思考。     

应用遗传神经网络方法分析我国降水化学数据

收集了１１１组包含Ｃａ＾２＋、Ｍｇ＾２＋等８种离子浓度和ＰＨ值的我国降水化学数据，并用遗传神经网络（Ｇｅｎｅｔｉｃ Ｎｅｕｒａｌ Ｎｅｔ－ｗｏｒｋｓ，ＧＮＮ）方法建立了这些离子浓度和ＰＨ值绵ＢＰ网络模式，运用这些模式计算了各离子浓度对ＰＨ值的贡献，结果表明，我国各地降水ＰＨ值与各地区酸性离子与碱性离子相互用途相关，其中生离子的中和作用起着主导作用。     

从环境类学术文献看中国环境教育存在的问题

根据我国环境科学类期刊上刊登的环境教育文献的比重和内容的调查结果来看，我国环境科学界对环境以重视程度与环境教育的重要性不成比例，环境教育研究的落后已成为制约我国环境教育可持续发展的关键问题。     

基于模糊聚类神经网络的煤炭资源分类研究

采用模糊聚类神经网络方法建立了煤炭资源分类模型 ,通过自适应算法与误差逆传播算法相结合 ,解决了以往神经网络不能处理模糊问题的缺陷 ,从而更适应现实评价条件的需要。     

水质评价模式识别的BP神经网络方法

水质受多个因素的影响，而且这些作用多是非线性的。其水质评价的实质是实现一个非线性映射。由于具有任意个隐层节点的前馈神经网络可以以任意精度逼近一个连续函数，因此，目前得到普遍应用的是采用BP算法的多层前馈神经网络。文中运用人工神经网络的反向传播（BP）算法进行了实例评价，与其它方法相比，评价精度较高，方法简单易行。     

神经网络技术在单机应变寿命监控中的应用研究

目的解决单机应变监控载荷模型中非线性问题带来的误差。方法基于人工神经网络技术,结合某型飞机典型盒段试验件有限元模型,建立用于单机应变寿命监控的神经网络载荷模型,利用随机选取的测试数据对模型的精确性进行验证,并与多元线性回归模型进行对比分析。结果 BP神经网络载荷模型的预测值比较贴近实测值,同时要优于多元线性回归模型的预测结果。结论 BP神经网络载荷模型可以用于单机应变监控,而且预测精度更高,可以更加准确地把握平尾的损伤情况。     

基于贝叶斯网络的河流突发性水质污染事故风险评估

事故造成的水质突发性风险评估对于河流水质安全管理具有重要意义.通过贝叶斯网络直观地表示事故风险源和河流水质之间的相关性,并用时序蒙特卡罗算法将风险源状态模拟、水质模拟和贝叶斯网络推理过程结合,可以对多个风险源共同影响下的河流突发性水质污染事故的超标风险进行量化评估.案例研究表明,多风险源对同一受纳水体的水质突发性污染事故风险的耦合影响十分显著,在进行流域水安全管理时必须对多风险源进行综合管理.同时,模型的诊断推理功能可为流域关键风险源识别和管理提供决策支持.