基于二分图理论的产业环境资源竞争网络模型研究——以广州开发区为例

采用二分图理论来研究区域内不同产业部门或排污企业之间对于环境容量的竞争关系,建立产业环境资源竞争网络模型,并从点权分布和加权集聚系数等方面加以分析,依据强关联矩阵来确定排污企业对于区域环境容量需求强度的聚类情况.在此基础上,进一步依据2010年广州开发区的排污企业数据进行了实证研究,分析了开发区产业现状,并对未来生态工业园区建设中产业的发展提出了指导性建议.采用复杂网络模型来分析企业环境外部性的方法,不仅较传统的经济度量方式有理论上创新,而且根据区域不同情况来构建模型,从复杂系统视角对环境管理体系进行分析,有利于改进环境管理方式.     

Characterization of biological responses under different environmental conditions: A hierarchical modeling approach

A natural river system is organized as a nested hierarchy of interconnected habitats with specific environmental conditions to which the biological community has adapted. Due to this hierarchical structure, identifying the role of different stressors on the biological community is a formidable task. Efforts trying to link stressors to biological integrity have always been bound to the geographic scale of the selected study area, leading to scale-specific results. In this research, an attempt is made to lift this limitation and develop a hierarchical, scale-sensitive methodology that can identify the significant environmental stressors to the biological community at different scales. Sites with similar background environmental conditions are clustered using self-organizing maps (SOM). This is used to identify stressors which affect the biological community throughout the area of study - called environmental gradients or large-scale stressors. Subsequently, these clusters of similar observations (sampling sites) are progressively sub-divided using environmental variables with a significant but localized effect on the biological community - called small-scale stressors. A parent group of sites is split only when the resulting sub-groups have significantly different biological responses. At the end of this recursive sites decomposition procedure, the original set of observations is organized as a tree of environmentally homogeneous groups of observations characterized by unique biological responses to multiple stressors with different geographic extents. The developed hierarchical analysis methodology has been validated using a large-size dataset of environmental observations from the State of Ohio. Our results show that habitat degradation and increased nutrient loading are the large-scale stressors with a widespread impact in Ohio. Other stressors, such as heavy metals, pH or nitrate concentrations have significant albeit localized effects on biological integrity.     

秋冬季区域性大气污染过程对长三角北部典型城市的影响

本文基于空气质量及气象监测数据,运用层次聚类分析法、后向轨迹法、潜在源贡献分析等方法,选取长三角北部地区2018~2019年秋冬季的典型污染过程进行分析,并选取该区域代表性的城市（蚌埠）进行深入分析.结果表明,长三角北部地区大气污染受地面弱气压、高湿、低温和静小风等不利气象条件及传输的影响较大.长三角秋冬季区域性污染具有影响范围广和持续时间长等特征,污染类型主要为区域外传输型与区域内累积型.在EP1和EP2两次污染过程中,长三角北部城市PM_2.5浓度均值分别达到131.6 μg·m^-3和115.4 μg·m^-3,前者污染过程较短,但污染物浓度累积较快造成的污染强度大和范围广.利用PSCF和CWT对PM_2.5潜在源定性和定量分析表明,EP1过程PM_2.5由临沂、徐州、宿迁和连云港等污染轨迹密集区域传输到受体城市蚌埠,CWT值处于80以上,最高可达200以上,区域传输实际浓度值较高;EP2过程PM_2.5浓度受宿迁、宿州和徐州等区域内部邻近城市影响,CWT值处于60以上,最高达160以上,说明研究区域内城市间相互影响较大.因此,秋冬季重污染过程中的跨区域大气污染联动,对于缓解长三角北部区域污染程度尤为重要.     

太原市城区夏季VOCs来源及其对O3生成的贡献

采集太原市城区夏季VOCs样品并分析其浓度特征,使用参数修正法得到VOCs初始浓度,分析其来源及对O₃生成的贡献.结果显示：太原市城区总VOCs平均浓度为48.13 μg/m³,烷烃（25.52 μg/m³）为主要组分.VOCs浓度呈明显日变化特征,在日间（10：00~14：00）光化学产生O₃的关键时段浓度最低.油品挥发、机动车排放、燃煤、植物排放与液化石油气/天燃气（LPG/NG）使用源对修正后环境VOCs的贡献分别为26.89%、25.55%、21.14%、14.99%、11.44%,对O₃生成的贡献分别为21.44%、33.10%、24.07%、13.77%、7.62%.机动车为新鲜排放气团VOCs的重要来源,而油品挥发、燃煤的输送与本地积累是其他（混合、夜间与反应）气团VOCs的重要来源.机动车排放、油品挥发与燃煤为VOCs与O₃生成的重要贡献源,控制此类源排放可减少太原市城区环境VOCs浓度并有效降低O₃生成.     

基于HYSPLIT4模式的天水市颗粒物输送路径

首先利用后向轨迹模式（HYSPLIT）模拟了天水市2017~2019年冬季后向轨迹,分析了50,200,500和1000m 4个不同高度以及500m高度不同聚类数量对路径输送聚类统计结果的影响,并以500m高度四个季节的后向轨迹进行聚类分析,进一步运用权重潜在源贡献分析法（WPSCF）,探讨了研究期间天水市细颗粒物的潜在源区及不同源区对天水市颗粒物浓度的贡献,结果表明,（1）起始点高度为500m时,颗粒物浓度极值比和极值差均较大,聚类结果最具代表性;（2）不同聚类数量分析结果表明,按照总空间变化（Total spatial variation,TSV）显著增加的原则选取的聚类数量较少,且能反映不同方向的轨迹输送特征;（3）天水市不同季节的轨迹聚类结果表明,冬季来自陕西南部的东南路径是PM_2.5污染程度最高的路径,该路径下PM_2.5浓度为78.2μg/m³,春季西北路径的颗粒物浓度最高,PM₁₀和PM_2.5的平均浓度分别是127.9~129.9和40.6~41.0 μg/m³,夏秋季节不同路径的颗粒物浓度相差不大.     

时延神经网络对宝兴树蛙异质种群鸣叫的聚类分析

对瓦屋山宝兴树蛙3个不同海拔分布(A:山下;B:山腰;C:山上)的异质种群鸣叫的时频参数进行分析,发现三地鸣叫各参数互相重叠,无明显界线.常规统计分析不能使三地的鸣叫聚类为各自的组.使用时延神经网络对三地宝兴树蛙鸣叫的时频参数进行有指导聚类,发现三种群鸣叫有地区性差异,并表现出由个体交流引起的山腰种群群体特征的不显著.推测差异可能是因为长期地理隔离所导致.基于神经网络的分析结果,对各种群间的个体交流情况进行了初步估计.图6表1参20     

系统加权聚类法及其在滑坡稳定性预测中的应用

本文以工程地质类比分析法为基础，针对一般系统聚类法在滑坡稳定性预测中应用的不足，提出并建立了一种系统加权聚类法。最后以三峡库岸研究程度较高的３６个边坡为例，进行了一般系统聚类法与系统加权聚类法的对比研究。对比研究结果表明：系统加权聚类法应用于滑坡稳定性预测，其预测精度有较大提高。     

A spatial clustering perspective on autocorrelation and regionalization

We revisit one of the classical problems in geography and cartography where multiple observations on a lattice (N) need to be grouped into many fewer regions (G), especially when this number of desired regions is unknown a priori. Since an optimization through all possible aggregations is not feasible, a hierarchical classification scheme is proposed with an objective function sensitive to spatial pattern. The objective function to be minimized during the assignment of observations to regions (classification) consists of two terms: the first characterizes accuracy and the second, model complexity. For the latter, we introduce a spatial measure that characterizes the number of homogeneous patches rather than the usual number of classes. A simulation study shows that such a classification procedure is less sensitive to random and spatially correlated error (noise) than non-spatial classification. We also show that for conditional autoregressive error (noise) fields the optimal partitioning is the one that has the highest within-units generalized Moran coefficient. The classifier is implemented in ArcView to demonstrate both a socio-economic and an environmental application to illustrate some potential applications.

Sustainable supplier selection criteria classification for Indian iron and steel industry: a fuzzy modified Kano model approach

ABSTRACT

The aim of this work is to provide a systematic approach for sustainability criteria clustering into different useful categories. For this purpose, a methodology has been proposed with the following three steps. (i) A comprehensive set of the sustainability criteria in economic, environmental and social dimension has been extracted from the literature and have been customised for the iron and steel industry. (ii) A fuzzy Kano methodology has been used to classify these sustainability criteria into useful Kano cluster like must-be, one-dimensional, and attractive cluster (iii) further, must-be and attractive criteria derived from the second step were further sub-classified into three categories using Type IV Kano model. All steps have been examined in the iron and steel industry of India and findings indicate that quality and delivery criteria are classified as high must be criteria in the economic realm. Further, environmental and energy management system in the environmental dimension and social responsibility and right of stakeholders criteria in the social dimension are classified as high must be criteria. Prioritising must be and attractive criteria enable decision makers of other industries to select the appropriate criteria to adopt sustainability.