A Procedure for Making Optimal Selection of Input Variables for Multivariate Environmental Classifications

Abstract:  Multivariate classifications of environmental factors are used as frameworks for conservation management. Although classification performance is likely to be sensitive to choice of input variables, these choices have been subjective in most previous studies. We used the Mantel test on a limited set of sites for which biological data were available to iteratively seek a definition of environmental space (i.e., intersite distances calculated with a set of appropriately transformed and weighted environmental variables) that had maximal correlation with the same sites described in a biological space. The procedure was used to select input variables for a classification of New Zealand's rivers that discriminates variation in fish communities for biodiversity management. The classification performed (i.e., discriminated biological variation) better than classifications with subjectively chosen variables. The inherently linear measures of environmental distance that underlie multivariate environmental classifications mean that they will perform best if they are defined based on variables for which there is a linear variation in the biological community throughout the entire range of the variable. Classification performance will therefore be improved when variables that have nonlinear relationships with biological variation are transformed to make their relationship with biological turnover more linear and when the contributions of environmental factors that have particularly strong relationships with biological variation are increased by weighting. Our results indicate that attention to the manner in which environmental space is defined improves the efficacy of multivariate classification and other techniques in which the environment is used as a surrogate for biological variation.     

Assessing the Threat Status of Ecological Communities

Abstract: Conservationists are increasingly interested in determining the threat status of ecological communities as a key part of their planning efforts. Such assessments are difficult because of conceptual challenges and a lack of generally accepted criteria. We reviewed 12 protocols for assessing the threat status of communities and identified conceptual and operational issues associated with developing a rigorous, transparent, and universal set of criteria for assessing communities, analogous to the International Union for Conservation of Nature (IUCN) Red List standards for species. We examined how each protocol defines a community and its extinction and how each applies 3 overarching criteria: decline in geographic distribution, restricted geographic distribution, and changes to ecological function. The protocols vary widely in threshold values used to assess declines and distribution size and the time frames used to assess declines, leading to inconsistent assessments of threat status. Few of the protocols specify a scale for measuring distribution size, although assessment outcomes are highly sensitive to scale. Protocols that apply different thresholds for species versus communities tend to require greater declines and more restricted distributions for communities than species to be listed in equivalent threat categories. Eleven of the protocols include a reduction in ecological function as a criterion, but almost all assess it qualitatively rather than quantitatively. We argue that criteria should be explicit and repeatable in their concepts, parameters, and scale, applicable to a broad range of communities, and address synergies between types of threats. Such criteria should focus on distribution size, declines in distribution, and changes to key ecological functions, with the latter based on workable proxies for assessing the severity, scope, and immediacy of degradation. Threat categories should be delimited by thresholds that are assessed at standard scales and are logically consistent with the viability of component species and important ecological functions.     

基于面向对象的北京市区城市内部用地信息提取

选取北京市四环以内为研究区域,以资源三号卫星遥感影像为数据源,针对类型多样、特征易混淆的城市内部用地以及高分辨率遥感影像海量信息、人工提取费时费力等特点,论文基于面向对象分类方法,探讨城市内部用地自动提取方法,并对分类结果进行精度评估.结果表明:利用不同地物的光谱、形状、纹理和空间关系等特征,通过多尺度分割和隶属度函数法,构建合理的分类层次,不仅精确提取出研究区内水体、绿地、建设用地和待开发用地,更独具创新地区分了城市建设用地内部各种地物类型,包括工业生产用地、低密度和高密度生活用地以及交通用地.该方法有效地利用了资源3 号卫星影像的光谱、纹理及空间信息特征,总体精度可达到87.00%,Kappa系数达到0.853 9,取得较好的分类效果.     

长江危险品运输风险下的分类管理方法研究

分析了危险品分类管理的现状及不足。在此基础上,研究了长江危险品运输的分类方法,并提出危险品分类管理的步骤,针对该方法在长江危险品运输的分类管理进行了实践。在危险品运输分类管理的基础上,提出了分类风险管理措施,以提高危险品运输的安全监管能力,改善长江危险品运输的安全监管结构,实现长江危险品运输的安全。     

南京城镇空间形态的度量和分析

进入20世纪80年代,中国的城市化进程开始加快,城市化使城镇空间结构发生迅速的变化,并伴随着城镇形态的变化。城镇形态的变化不仅反映了城镇发展的方向、发展规模的变化,也反映了城镇与自然环境的关系,揭示了城镇发展与区域空间结构的关系。对城镇形态的定量化研究不仅深化城镇空间结构和城镇化的理论研究,而且有助于城镇规划的科学性,包括城镇的发展方向、发展规模、发展速度的优化和调控等。主要以南京市域城镇为案例,利用遥感技术的仿归一化植被指数法,提取2000年9月份南京Landsat TM影像上的城镇空间信 息,并用定量的方法,对城镇形态进行度量、分类,最后根据分类结果分析了城镇形态形成的背景和机制。据此,为城镇化进程中的城镇发展规划提供科学的参考依据。     

高处坠落事故的人因失误与干预策略研究

为系统分析导致高处坠落人因事故的产生机理,通过统计152起建筑工程高处坠落事故的调查与分析报告,从组织影响、安全监管、不安全行为前提条件和不安全行为等4个层次,辨识影响高处坠落事故的人为失误因素,修订人为因素分析与分类系统框架（HFACS）。设计高处坠落人因失误调查问卷,开展一线高处作业人员问卷调查,建立高处坠落人因失误结构方程模型,对导致高处坠落事故的人为失误因素进行路径分析。结果表明:各潜在因素间均呈正相关,且高处坠落人因失误事故的关键路径为资源管理不到位→安全监督培训不充分→班组管理不良→操作违规。综合各因素间相关性,提出了针对性的预防高处坠落事故的人因干预策略。     

煤矿安全隐患动态分级闭环管理方法及应用

为了治理煤矿生产过程中的安全隐患,遏制煤矿安全生产事故,对煤矿生产过程中的安全隐患管理进行了研究。经过梳理国内外相关文献,辨析了煤矿危险源和安全隐患的关系;结合脆弱性分析了煤矿安全隐患的风险大小,结合LEC法说明了安全隐患风险分级标准;结合煤矿生产过程的特点,提出了煤矿安全隐患动态分级闭环管理方法,设计了与之配套的奖惩办法。煤矿安全隐患动态分级闭环管理方法和配套的奖惩办法在南方的一个千万吨级矿井进行了应用。研究表明,该方法可以有效提高煤矿安全隐患的治理能力,提高煤矿现场的安全管理水平。     

中国西北地区农林复合经营的保护与发展

中国西北地区深居内陆,区域生态系统脆弱,干旱、水土流失和风沙灾害等问题频繁,严重影响了区域社会经济的发展与生态文明建设进程。发展复合农林业是林业发展的新思路,是解决区域生态环境与落后社会经济问题恶性循环的有效途径。论文根据西北地区特殊的自然、经济、社会条件以及农林复合经营发展现状,结合农林复合系统科研、生产实践及农村经济发展的需要,将西北干旱地区农林复合经营划分为草原风沙区、荒漠绿洲区、黄土高原区和青藏高原区4个类型区,并分析概述了各类型区复合农林业发展的现状和特点。根据农林复合系统多层次、多组分、多功能和多目标的特点,提出了基于林分起源、生物组分、生长周期和主导功能标尺的农林复合经营类型划分方法。对目前西北地区农林复合经营发展中存在的规划、理论研究和推广应用等问题提出了相应对策,以期为区域农林复合经营的保护与发展提供借鉴。     

1种基于视频的油田危险区域入侵检测智能综合识别技术研究

为解决油田日常生产作业中缺乏危险作业区域的等级划分与自动识别方式以及缺乏人员踏入危险区域的识别方法.提出1种基于视频智能综合识别技术的全天油田危险区域入侵检测算法,该算法首先结合油田危险因素对油田危险区域进行危险等级的划分与危险区域的识别;然后,针对光照条件良好的白天场景,在训练数据集中融合油田作业区视频数据和公开行人...     

卫星云图样本集的FCM优化调整与云类判别

基于云类样本的红外-可见光二维灰度空间投影,采用模糊聚类方法调整优化云类样本特征区域,消除采样误差。针对常规模糊C均值聚类(FCM)方法在处理上述问题时表现出的局限性,提出用样本特征均值替代FCM中随机初始中心的改进办法,既避免了常规FCM方法对初始中心敏感的缺陷,又可纠正其聚类结果对云类样本特征结构的歪曲。改进后的聚类结果既消除了采样误差,又保持了云类样本的基本特征属性,基于该判据的分类结果可较为准确地分辨出陆地、水体、低云、中云、卷云和积雨云,分割判别效果符合客观实际。