遗传投影寻踪回归在城市环境质量评价中的应用

引入投影降维的思想,将遗传投影寻踪与回归分析技术运用到城市环境质量评价中。将此技术与神经网络方法进行实例比较,投影寻踪回归方法不但可以合理地作出环境质量的综合评价,而且消除了神经网络方法中类别判断不够精确的影响。     

Self-Organizing Maps for Integrated Environmental Assessment of the Mid-Atlantic Region

A new method has been developed to perform environmental assessment at regional scale. This involves a combination of a self-organizing map (SOM) neural network and principal component analysis (PCA). The method is capable of clustering ecosystems in terms of environmental conditions and suggesting relative cumulative environmental impacts of multiple factors across a large region. Using data on land-cover, population, roads, streams, air pollution, and topography of the Mid-Atlantic region, the method was able to indicate areas that are in relatively poor environmental condition or vulnerable to future deterioration. Combining the strengths of SOM with those of PCA, the method offers an easy and useful way to perform a regional environmental assessment. Compared with traditional clustering and ranking approaches, the described method has considerable advantages, such as providing a valuable means for visualizing complex multidimensional environmental data at multiple scales and offering a single assessment or ranking needed for a regional environmental assessment while still facilitating the opportunity for more detailed analyses.     

基于ANN的山东省可持续发展水平的区域差异

针对区域可持续发展系统的非线性,采用人工神经网络中较为先进的自组织特征映射网络构建模型.以山东省17个地市2001年的经济、人口、资源、环境、生活和科教状况作为带分样本,用自组织特征映射模型对山东省可持续发展区域差异进行了研究.研究结果表明,山东省可持续发展水平的区域差异可划分为4类,模拟结果比较理想.     

Assessment of groundwater quality by means of self-organizing maps: application in a semiarid area

The Kohonen neural network was applied to hydrochemical data from the Detritic Aquifer of the Lower Andarax, situated in a semiarid zone in the southeast of Spain. An activation map was obtained for each of the sampling points, in which the spatial distribution of the activated neurons indicated different water qualities. To extract the information contained in the activation maps, they were divided into nine quadrats. Cartesian coordinates were assigned to each quadrant (x, y), and for each sampling point, three derived variables were selected, which were assigned the values x and y of the corresponding quadrat. A classification was defined based on this simple matrix system which allows an easy and rapid means of evaluating the water quality. This assessment highlights the various processes that affect groundwater quality. The method generates output that is easier to interpret than from traditional statistical methods. The information is extracted from the activation maps without significant loss of information. The method is proposed for assessing water quality in hydrogeochemically complex areas, where large numbers of observations are made.     

Synergistic Techniques for Better Understanding and Classifying the Environmental Structure of Landscapes

The desire to capture natural regions in the landscape has been a goal of geographic and environmental classification and ecological land classification (ELC) for decades. Since the increased adoption of data-centric, multivariate, computational methods, the search for natural regions has become the search for the best classification that optimally trades off classification complexity for class homogeneity. In this study, three techniques are investigated for their ability to find the best classification of the physical environments of the Mt. Lofty Ranges in South Australia: AutoClass-C (a Bayesian classifier), a Kohonen Self-Organising Map neural network, and a k-means classifier with homogeneity analysis. AutoClass-C is specifically designed to find the classification that optimally trades off classification complexity for class homogeneity. However, AutoClass analysis was not found to be assumption-free because it was very sensitive to the user-specified level of relative error of input data. The AutoClass results suggest that there may be no way of finding the best classification without making critical assumptions as to the level of class heterogeneity acceptable in the classification when using continuous environmental data. Therefore, rather than relying on adjusting abstract parameters to arrive at a classification of suitable complexity, it is better to quantify and visualize the data structure and the relationship between classification complexity and class homogeneity. Individually and when integrated, the Self-Organizing Map and k-means classification with homogeneity analysis techniques also used in this study facilitate this and provide information upon which the decision of the scale of classification can be made. It is argued that instead of searching for the elusive classification of natural regions in the landscape, it is much better to understand and visualize the environmental structure of the landscape and to use this knowledge to select the best ELC at the required scale of analysis.     

基于深度学习的大数据空气污染预报

为了更好地反映环境污染变化趋势,为环境管理决策提供及时、全面的环境质量信息,预防严重污染事件发生,开展城市空气质量预报研究是十分必要的.本文针对环境大数据时代下的城市空气质量预报,提出了一种基于深度学习的新方法.该方法通过模拟人类大脑的神经连接结构,将数据在原空间的特征表示转换到具有语义特征的新特征空间,自动地学习得到层次化的特征表示,从而提高预报性能.得益于这种方式,新方法与传统方法相比,不仅可以利用空气质量监测、气象监测及预报等环境大数据,充分考虑污染物的时空变化、空间分布,得到语义性的污染物变化规律,还可以基于其他空气污染预测方法的结果(如数值预报模式),自动分析其适用范围、优势劣势.因此,新方法通过模拟人脑思考过程实现更充分的大数据集成,一定程度上克服了现有方法的缺陷,应用上更加具有灵活性和可操作性.最后,通过实验证明新方法可以提高空气污染预报性能.     

LAKE WATER QUALITY ASSESSMENT FROM LANDSAT THEMATIC MAPPER DATA USING NEURAL NETWORK: AN APPROACH TO OPTIMAL BAND COMBINATION SELECTION1

Abstract: The concern about water quality in inland water bodies such as lakes and reservoirs has been increasing. Owing to the complexity associated with field collection of water quality samples and subsequent laboratory analyses, scientists and researchers have employed remote sensing techniques for water quality information retrieval. Due to the limitations of linear regression methods, many researchers have employed the artificial neural network (ANN) technique to decorrelate satellite data in order to assess water quality. In this paper, we propose a method that establishes the output sensitivity toward changes in the individual input reflectance channels while modeling water quality from remote sensing data collected by Landsat thematic mapper (TM). From the sensitivity, a hypothesis about the importance of each band can be made and used as a guideline to select appropriate input variables (band combination) for ANN models based on the principle of parsimony for water quality retrieval. The approach is illustrated through a case study of Beaver Reservoir in Arkansas, USA. The results of the case study are highly promising and validate the input selection procedure outlined in this paper. The results indicate that this approach could significantly reduce the effort and computational time required to develop an ANN water quality model.     

基于人工神经网络的突发环境事故应急系统

突发环境事故现场具有形势紧迫、信息获取困难等特点,引入人工神经网络技术(Artificial Neural Networks,ANN)设计开发的突发环境事故应急决策支持系统(ANN-EEDSS),能够完成现场应急指挥调度的定量化决策支持。关键技术包括突发环境事故神经网络构建方法,事故信息与应急措施的数字化方法,决策信息的翻译表示,事故影响因素的权重分析。最后阐述了系统现场运行和工作的流程。     

Are 'other gainful activities' on farms good for the environment?

There has been a lot of academic interest in the pursuit of diversification activities and off-farm employment by farm household members. This is regarded as an important strategy for mitigating the effects of low agricultural income. One aspect of the debate about these so-called 'Other Gainful Activities' (OGAs) is whether they are associated with any environmental improvement on farms. In this paper, we use three approaches to analyse this issue. We examine whether measures of agricultural intensity are associated with the pursuit of OGAs by farmers and their spouses. We examine whether OGAs are more likely on farms where there is an ESA Management Agreement. Finally, we examine whether OGAs are associated with the farmer's stated environmental intentions. Although we tentatively conclude that there is a relationship between OGA involvement and these measures of environmental performance or concern by farmers, the magnitude of the association is small relative to other variables such as farm size, the type of land use, the form of business and recent agricultural training.     

Analyzing Human–Landscape Interactions: Tools That Integrate

Humans have transformed much of Earth’s land surface, giving rise to loss of biodiversity, climate change, and a host of other environmental issues that are affecting human and biophysical systems in unexpected ways. To confront these problems, environmental managers must consider human and landscape systems in integrated ways. This means making use of data obtained from a broad range of methods (e.g., sensors, surveys), while taking into account new findings from the social and biophysical science literatures. New integrative methods (including data fusion, simulation modeling, and participatory approaches) have emerged in recent years to address these challenges, and to allow analysts to provide information that links qualitative and quantitative elements for policymakers. This paper brings attention to these emergent tools while providing an overview of the tools currently in use for analysis of human–landscape interactions. Analysts are now faced with a staggering array of approaches in the human–landscape literature—in an attempt to bring increased clarity to the field, we identify the relative strengths of each tool, and provide guidance to analysts on the areas to which each tool is best applied. We discuss four broad categories of tools: statistical methods (including survival analysis, multi-level modeling, and Bayesian approaches), GIS and spatial analysis methods, simulation approaches (including cellular automata, agent-based modeling, and participatory modeling), and mixed-method techniques (such as alternative futures modeling and integrated assessment). For each tool, we offer an example from the literature of its application in human–landscape research. Among these tools, participatory approaches are gaining prominence for analysts to make the broadest possible array of information available to researchers, environmental managers, and policymakers. Further development of new approaches of data fusion and integration across sites or disciplines pose an important challenge for future work in integrating human and landscape components.     

大气环境质量分析与评价研究的新进展

简要地介绍了人工神经网络（ANN）、物元可拓集（MEES）、集对分析（SPA）、遗传算法（GA）和投影寻踪技术（PP）等的基本思想和特点，评述了近年来国内外将这些新理论和新技术应用于大气环境质量分析与评价取得的最新成果，指出了大气环境质量分析与评价研究的可能发展前景。     

Flood Forecasting Using Neural Computing Techniques and Conceptual Class Segregation

Various neural networks models are developed and applied for flood forecasting at Sangye station (no. 1) of the Bocheong Stream catchment, which is one of the International Hydrological Program's representative catchments, Republic of Korea. The neural networks models (NNMs) are multilayer perceptron‐neural networks model (MLP‐NNM), generalized regression neural networks model (GRNNM), and Kohonen self‐organizing feature maps neural networks model (KSOFM‐NNM). Data used for model training and testing are divided into two groups: such as floods and typhoon events. Single conventional application and class segregation implementation are applied to evaluate the neural networks models. KSOFM‐NNM forecasts flood discharge more accurately than do MLP‐NNM and GRNNM for the testing data of Methods I and II for single conventional application and class segregation implementation. This study shows that class segregation can capture the dynamics of different physical processes and overcome the difficulties using single conventional application of neural networks models.     

Cluster analysis of structural stage classes to map wildland fuels in a Madrean ecosystem

Geospatial information technology is changing the nature of fire mapping science and management. Geographic information systems (GIS) and global positioning system technology coupled with remotely sensed data provide powerful tools for mapping, assessing, and understanding the complex spatial phenomena of wildland fuels and fire hazard. The effectiveness of these technologies for fire management still depends on good baseline fuels data since techniques have yet to be developed to directly interrogate understory fuels with remotely sensed data. We couple field data collections with GIS, remote sensing, and hierarchical clustering to characterize and map the variability of wildland fuels within and across vegetation types. One hundred fifty six fuel plots were sampled in eight vegetation types ranging in elevation from 1150 to 2600 m surrounding a Madrean 'sky island' mountain range in the southwestern US. Fuel plots within individual vegetation types were divided into classes representing various stages of structural development with unique fuel load characteristics using a hierarchical clustering method. Two Landsat satellite images were then classified into vegetation/fuel classes using a hybrid unsupervised/supervised approach. A back-classification accuracy assessment, which uses the same pixels to test as used to train the classifier, produced an overall Kappa of 50% for the vegetation/fuels map. The map with fuel classes within vegetation type collapsed into single classes was verified with an independent dataset, yielding an overall Kappa of 80%.     

Cancer atlases: Uses and limitations

Summary Atlases of cancer mortality have recently appeared in many nations. This paper cautions environmental professionals against using them to draw cause-and-effect conclusions because of limitations of the map as a conveyor of information, and limitations of the data used in cancer atlases. Appropriate uses include, finding clues for detailed epidemiological and environmental research, places with poor data, regions requiring special equipment and personnel, and as tools for persuading the public to be less fatalistic about cancer and more supportive of public health protection and education programs.Dr Michael Greenberg is both Professor of the School of Urban and Regional Policy, and Acting Director of the Graduate Program in Public Health, at the University of Medicine and Dentistry and Rutgers University.     

Representative Landscapes in the Forested Area of Canada

Canada is a large nation with forested ecosystems that occupy over 60% of the national land base, and knowledge of the patterns of Canada’s land cover is important to proper environmental management of this vast resource. To this end, a circa 2000 Landsat-derived land cover map of the forested ecosystems of Canada has created a new window into understanding the composition and configuration of land cover patterns in forested Canada. Strategies for summarizing such large expanses of land cover are increasingly important, as land managers work to study and preserve distinctive areas, as well as to identify representative examples of current land-cover and land-use assemblages. Meanwhile, the development of extremely efficient clustering algorithms has become increasingly important in the world of computer science, in which billions of pieces of information on the internet are continually sifted for meaning for a vast variety of applications. One recently developed clustering algorithm quickly groups large numbers of items of any type in a given data set while simultaneously selecting a representative—or “exemplar”—from each cluster. In this context, the availability of both advanced data processing methods and a nationally available set of landscape metrics presents an opportunity to identify sets of representative landscapes to better understand landscape pattern, variation, and distribution across the forested area of Canada. In this research, we first identify and provide context for a small, interpretable set of exemplar landscapes that objectively represent land cover in each of Canada’s ten forested ecozones. Then, we demonstrate how this approach can be used to identify flagship and satellite long-term study areas inside and outside protected areas in the province of Ontario. These applications aid our understanding of Canada’s forest while augmenting its management toolbox, and may signal a broad range of applications for this versatile approach.     

Advanced in situ spectroscopic techniques and their applications in environmental biogeochemistry: introduction to the special section

Understanding the molecular-scale complexities and interplay of chemical and biological processes of contaminants at solid, liquid, and gas interfaces is a fundamental and crucial element to enhance our understanding of anthropogenic environmental impacts. The ability to describe the complexity of environmental biogeochemical reaction mechanisms relies on our analytical ability through the application and developmemnt of advanced spectroscopic techniques. Accompanying this introductory article are nine papers that either review advanced in situ spectroscopic methods or present original research utilizing these techniques. This collection of articles summarizes the challenges facing environmental biogeochemistry, highlights the recent advances and scientific gaps, and provides an outlook into future research that may benefit from the use of in situ spectroscopic approaches. The use of synchrotron-based techniques and other methods are discussed in detail, as is the importance to integrate multiple analytical approaches to confirm results of complementary procedures or to fill data gaps. We also argue that future direction in research will be driven, in addition to recent analytical developments, by emerging factors such as the need for risk assessment of new materials (i.e., nanotechnologies) and the realization that biogeochemical processes need to be investigated in situ under environmentally relevant conditions.     

基于投影寻踪动态聚类模型的边坡稳定性评价

投影寻踪聚类(PPC)模型在多因素聚类分析中被广泛应用并取得了满意的效果,然而,该模型还存在诸如密度窗宽参数取值经验确定等不足,有待改进提高。本文针对PPC模型存在的问题,把投影寻踪聚类的思想和动态聚类方法结合起来构造投影指标,建立了投影寻踪动态聚类(PPDC)新模型,新模型在整个运算过程中毋需人为给定参数,聚类结果客观、明确。边坡稳定性评价的实际应用表明,PPDC模型不仅切实可行,取得了理想的效果,而且还具有稳定性好、操作简便等特点。PPDC模型为多因素聚类分析提供了一种新方法,具有广阔的应用前景。     

A collaborative approach to address the cumulative impacts of mine-water discharge: Negotiating a cross-sectoral waterway partnership in the Bowen Basin,Australia

The social and environmental impacts of rapidly expanding coal and gas industries have generated high levels of public concern and there is increasing evidence of cumulative impacts. In the Bowen Basin of Queensland (Australia) water quality issues have triggered a collaborative response to coordinate monitoring efforts, integrate data and information and undertake regional analysis to inform landscape-scale management. Collaborative governance is promoted as a response to complex environmental problems, such as cumulative impacts. However, application of this approach to the resources and energy sectors remains a significant research gap. This paper reports the results of action research in the 2 years taken to negotiate the establishment of collaborative governance arrangements to address mine-water discharge impacts in the Bowen Basin. The long establishment phase has been required to refine objectives, build trust, develop governance mechanisms and secure resourcing commitments. The partnership established involves more than 20 organisations including regulators, resources and energy companies, agricultural industries and research organisations. The breadth of participating sectors is a significant innovation, but also represents a major challenge in establishing this model of regional environmental governance. Promising strategies adopted to manage these tensions have included neutral brokerage, facilitative leadership, establishing legitimacy of the collaboration and credibility of its reports. The case study provides a cautionary tale of the pursuit of the promise of ‘everyone working together’ to address cumulative impacts. Policy implications include the need for extended commitment and integration of collaborative and other responses.     

基于多模态感知的危险废物环境风险评估与预警研究

为了有效地防范化解危险废物的环境风险,亟须构建风险的精细化评估与实时智能预警系统。然而,危险废物特性复杂、全过程流转链条长,传统的单一感知数据并不能满足环境风险评估和预警的需求。因此,本文提出基于多模态感知的危险废物环境风险评估与预警方法,采用物联网、大数据、人工智能等技术手段,实现危险废物基础信息、环境要素、空间信息的多模态实时感知,并通过感知信息的融合、分析,进行准则判断或现象描述。围绕以上研究思路,本文针对危险废物贮存环节及非法倾倒两个重要风险因子释放情景,设计了基于多模态感知的风险评估与预警方法,以期为危险废物环境风险的管控提供相应的服务与决策。     

Vulnerability and resilience to climate change in Sorsogon City,the Philippines: learning from an ordinary city?

Sorsogon City is a rapidly urbanising coastal area in the Philippines. Its location, combined with a rapidly changing and growing urban fabric, leaves it vulnerable to both incremental climate change and associated extreme weather events. In this paper, UN-HABITAT data are used to draw out the climate change vulnerabilities and policy responses in Sorsogon City. Vulnerability “hotspots” highlight the spatial intersection of socio-economic justice concerns, particularly in terms of vulnerability to increased cyclone activity. We discuss vulnerabilities of Sorsogon City and its citizens to climate change and measures undertaken through various social, environmental and technical systems and interventions to increase resilience. The paper also attempts to unpick the relationship between the neat, concise reported city and the complexities of urban life using the Sorsogon experiment to consider the limitations of such approaches to governing climate change. We group these under four headings: social simplification in the absence of data; over-governance (and under-representation); quick wins versus strategic investment; and stretching the ecological and vulnerability footprints of the city. The experience of Sorsogon City is then extended to reflect on issues of governance and planning in other Asian coastal cities.