Mapping wetlands in the Lower Mekong Basin for wetland resource and conservation management using Landsat ETM images and field survey data

The Mekong River Basin is considered to be the second most species rich river basin in the world. The 795,000 km(2) catchment encompasses several ecoregions, incorporating biodiverse and productive wetland systems. Eighty percent of the rapidly expanding population of the Lower Mekong Basin (LMB), made up in part by Lao PDR, Thailand, Cambodia and Viet Nam, live in rural areas and are heavily reliant on wetland resources. As the populations of Cambodia and Lao PDR will double in the next 20 years, pressure on natural resources and particularly wetlands can only increase. For development planning, resource and conservation management to incorporate wetland issues, information on the distribution and character of Mekong wetlands is essential. The existing but outdated wetland maps were compiled from secondary landuse-landcover data, have limited coverage, poor thematic accuracy and no meta-data. Therefore the Mekong River Commission (MRC) undertook to produce new wetland coverage for the LMB. As resources, funding and regional capacity are limited, it was determined that the method applied should use existing facilities, be easily adaptable, and replicable locally. For the product to be useful it must be accepted by local governments and decision makers. The results must be of acceptable accuracy (>75%) and the methodology should be relatively understandable to non-experts. In the first stage of this exercise, field survey was conducted at five pilot sites covering a range of typical wetland habitats (MRC wetland classification) to supply data for a supervised classification of Landsat ETM images from the existing MRC archive. Images were analysed using ERDAS IMAGINE and applying Maximum Likelihood Classification. Field data were reserved to apply formal accuracy assessment to the final wetland habitat maps, with resulting accuracy ranging from 77 to 94%. The maps produced are now in use at a Provincial and National level in three countries for resource and conservation planning and management applications, including designation of a Ramsar wetland site of international importance.     

Building spectral libraries for wetlands land cover classification and hyperspectral remote sensing

Recent advances in remote sensing provide opportunities to map plant species and vegetation within wetlands at management relevant scales and resolutions. Hyperspectral imagers, currently available on airborne platforms, provide increased spectral resolution over existing space-based sensors that can document detailed information on the distribution of vegetation community types, and sometimes species. Development of spectral libraries of wetland species is a key component needed to facilitate advanced analytical techniques to monitor wetlands. Canopy and leaf spectra at five sites in California, Texas, and Mississippi were sampled to create a common spectral library for mapping wetlands from remotely sensed data. An extensive library of spectra (n=1336) for coastal wetland communities, across a range of bioclimatic, edaphic, and disturbance conditions were measured. The wetland spectral libraries were used to classify and delineate vegetation at a separate location, the Pacheco Creek wetland in the Sacramento Delta, California, using a PROBE-1 airborne hyperspectral data set (5m pixel resolution, 128 bands). This study discusses sampling and collection methodologies for building libraries, and illustrates the potential of advanced sensors to map wetland composition. The importance of developing comprehensive wetland spectral libraries, across diverse ecosystems is highlighted. In tandem with improved analytical tools these libraries provide a physical basis for interpretation that is less subject to conditions of specific data sets. To facilitate a global approach to the application of hyperspectral imagers to mapping wetlands, we suggest that criteria for and compilation of wetland spectral libraries should proceed today in anticipation of the wider availability and eventual space-based deployment of advanced hyperspectral high spatial resolution sensors.     

地震灾区分级和灾害程度排序方法研究——以汶川8.0级地震为例

根据汶川8.0级地震震害评估工作的具体实践,探讨了地震灾区分级和灾害程度排序的方法。通过房屋震害系数、强震加速度观测记录、发震构造、地震地质灾害和场地条件、人口分布及伤亡情况等影响因素来确定受灾程度,以烈度区、统计学方法、速报灾情等来确定灾区分级。     

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。     

基于支持向量机的化工工艺安全评价模型构建及优化研究

为解决现阶段基于风险分级的安全评价方法仍存在着高维数据处理不当、评价智能化程度不高等问题,创建支持向量机的安全评价模型,利用核函数解决安全评价因子分类问题,粒子群算法(PSO)寻找最适合模型的正则项C,进一步提升安全评价模型的正确率,形成适用高维数据的化工工艺安全评价方法。研究结果表明：该模型与经典支持向量机模型和BP神经网络评价模型相比具有更高的正确率,研究结果对借用机器学习来创新安全评价理论及工程应用具有现实意义及理论价值。     

中国森林火灾研究综述

概述了中国森林火灾方面的研究成果,包括森林火灾的分级和分类、影响因素、研究内容、预报方法等,重点论述了中国森林火灾的研究内容,尤其是中国森林火灾的发生规律及学者们对森林火灾原因的看法。最后提出了防治森林火灾的指导思想。     

广州市大气气溶胶散射吸湿增长因子的观测研究

利用环境湿度的日变化,对广州市大气气溶胶散射吸湿增长因子进行了观测研究.基于反向轨迹模型HYSPLIT,将2006年7月到达广州市区的气溶胶分为海洋型、海洋/城市混合型、城市型3个类型.整个观测期间,海洋型、海洋/城市混合型、城市型气溶胶的散射吸湿增长因子f(RH=80%)的平均值分别为2.68±0.59、2.29±0.28、2.04±0.28.经验公式f(RH)=1+a(RH/100)b能够较好地拟合3种类型的气溶胶散射吸湿增长因子,海洋型、海洋城市混合型、城市型气溶胶的散射吸湿增长因子的拟合经验参数a分别为4.92、3.26、2.06,经验参数b分别为5.04、3.85、3.60.     

鄱阳湖地区洪水灾害模式分析

在对多年水文数据进行分析的基础上，利用Kohonen自组织映射对鄱阳湖地区降水时序变化、长江来水和湖水水位变化进行模式识别。之后，将所得的模式作为解释变量，用CART方法对样本年份进行分类，建立了鄱阳湖区灾害模式判别树。借此揭示了鄱阳湖地区降水和长江来水对鄱阳湖洪涝灾害的影响。     

基于支持向量机的水资源安全评价

支持向量机以统计学习理论为基础,采用结构风险最小化准则,将学习问题转化为一个凸二次规划问题,能够得到全局最优解,适合解决小样本、非线性分类及回归问题。根据水资源安全的内涵,筛选出具有代表性的指标,组成水资源安全评价指标体系。建立了基于支持向量机的水资源安全评价模型,将安全标准划分为良好、安全、临界、不安全、危险5个等级。根据水资源安全评价标准及所属评价等级值,随机生成样本集,180个样本作为训练样本,构造了5个两类支持向量分类器,20个样本作为检验样本,检验样本分类全部正确。将模型应用于山西省11个城市的水资源安全评价,结果表明,该方法是有效、可行的。     

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。