Landslide susceptibility zonation method based on C5.0 decision tree and K-means cluster algorithms to improve the efficiency of risk management

Machine learning algorithms are an important measure with which to perform landslide susceptibility assessments,but most studies use GIS-based classification methods to conduct susceptibility zonation.This study presents a machine learning approach based on the C5.0 decision tree(DT)model and the K-means cluster algorithm to produce a regional landslide susceptibility map.Yanchang County,a typical landslide-prone area located in northwestern China,was taken as the area of interest to introduce the proposed application procedure.A landslide inventory containing 82 landslides was prepared and subse-quently randomly partitioned into two subsets:training data(70％landslide pixels)and validation data(30％landslide pixels).Fourteen landslide influencing factors were considered in the input dataset and were used to calculate the landslide occurrence probability based on the C5.0 decision tree model.Susceptibility zonation was implemented according to the cut-off values calculated by the K-means clus-ter algorithm.The validation results of the model performance analysis showed that the AUC(area under the receiver operating characteristic(ROC)curve)of the proposed model was the highest,reaching 0.88,compared with traditional models(support vector machine(SVM)=0.85,Bayesian network(BN)=0.81,frequency ratio(FR)=0.75,weight of evidence(WOE)=0.76).The landslide frequency ratio and fre-quency density of the high susceptibility zones were 6.76/km2 and 0.88/km2,respectively,which were much higher than those of the low susceptibility zones.The top 20％interval of landslide occurrence probability contained 89％of the historical landslides but only accounted for 10.3％of the total area.Our results indicate that the distribution of high susceptibility zones was more focused without contain-ing more"stable"pixels.Therefore,the obtained susceptibility map is suitable for application to landslide risk management practices.     

四川盆地云和降水观测科学试验设计与实践

2014—2017年四川地区开展了大范围云系观测科学试验。观测对象以盆地层状云系和积层混合云系为主,积状云(对流云)为辅。本文围绕试验目标、区域、观测要素、观测布局设计、观测方案设计、设备技术参数和典型个例等7个方面进行介绍,根据不同类型云系和降水变化特征,设计有针对性的观测方案,获得了不同类型云系和降水的多尺度连续性观测数据,为四川地区开展云和降水关系研究提供详实的综合型外场观测数据。层状降水云典型个例云高可达8km,云强核心部位的云雷达反射率可达28dBz,径向速度可达-6m·s~(-1),在0℃附近,反射率和退偏振因子LDR上有一条明显的亮带,表现为极大值,液态水主要集中于4.5km以下,随着雨强增大,液水含量增加,降水滴谱分布较窄,随着雨强减小,雨滴谱和速度谱变窄,但小粒径数浓度增加,说明对层状云雨强起主导作用的是雨滴直径,而不是数浓度;无降水层状云典型个例云层厚度为3.2km,云顶约为4km,云雷达反射率不超过0dBz,径向速度不超过5m·s~(-1),层状云内整层水汽含量和液水含量较为稳定,云中主要为液态粒子且粒径偏小、小粒径数浓度较高。     

基于位置微博的大风信息提取及应用

利用广东省广州市2017年1月1日至5月30日的大风信息和灾情数据,分析了大城市位置微博的时空特征,在此基础上采用关键词语意分析、风险等级映射、灾情权重值结合GIS热点分析的方法构建基于位置微博的大风信息提取架构,实现大风信息的提取,并利用个例进行验证。研究表明:本文提出基于位置微博的大风信息提取架构,对收集大风信息和灾情具有一定的实用价值。而位置微博信息本身具有及时性强、成本低、数据量大的优势,能作为一种大风信息提取的有效补充手段。     

基于小波长短期记忆网络的风电功率超短期概率预测

随着大规模的风电并网,风电所具有的间歇性与随机性对电力系统的稳定性产生了很大的影响,风电功率预测成为当前解决该问题重要的方式之一.本文利用长短期记忆（LSTM）网络良好的时序记忆特性,将小波分解技术与LSTM深度网络结合,提出基于小波长短期记忆网络的风电功率超短期概率预测模型.首先通过小波分解技术将原始时间序列进行平稳化处理,再建立各子序列样本的LSTM网络预测模型,借助最大似然估计法估计预测误差的高斯分布函数,最终实现对未来4 h时刻的风电功率概率区间预测.最后,采用中国东北某风电场数据对所提方法进行算例分析,结果表明,将小波分解与深度学习方法结合可以较好地提高预测的精度,提高概率预测的区间可靠性.     

Geochemical characteristics of Early Cretaceous source rocks in Boli Basin

The Early Cretaceous deposits are composed of important source rocks in Boli Basin. The types of the source rocks include black mudstones and coal (with carbonaceous mudstone). By the organic geochemical analysis methods, the authors discussed the organic petrological characters, abundance of organic matter, degree of maturity and the type of source rocks. The main micro-component of black mudstone is exinite or vitrinite, and the content of vitrinite is high in coal. The weathering of the outcrop is very serious. The abundance of organic matter in source rock reaches the poor to better rank. The major kerogens in mudstone are type-Ⅲ, type-Ⅱ2 and some type-Ⅱ1; the organic type of coal is type-Ⅲ. The thermal evolution of the source rocks in every structural unit is very different, from low-maturity to over-maturity. The depositional environment is reductive, which is good for the preservation of organic matter. The organic matter in source rocks is mainly from aquatic organisms and terrigenous input.     

珠江三角洲城市旅游竞争力空间结构体系初探

区域城市间旅游竞争力空间结构体系是一个由不同等级、不同空间区位、不同旅游发展阶段的城市通过旅游资源、旅游交通等生产力要素以及客流有机耦合而形成的。本文在深入分析旅游竞争力含义及其空间结构的基础上,采用城市旅游竞争力的多维评价和综合分析的方法,将珠江三角洲城市旅游目的地划分为不同的3个等级,并总结出珠江三角洲城市旅游竞争力具有以广深为轴线的“哑铃”型双极放射状空间形态模式,提出了凭借城市自身比较优势,承担不同层次旅游竞争与合作角色的空间组织模式,旨在通过研究梳理各城市旅游影响要素的优劣差异,认识区域内旅游竞争力的空间结构特征,为进一步探讨形成互利共赢的城市旅游竞争与合作机制奠定基础,促进珠江三角洲城市旅游经济的和谐发展。     

中更新世气候转型时期南海生态环境的南北差异

中更新世气候转型在南海浮游有孔虫、氧同位素和其它生物记录上主要反映在900 ka BP前后发生高频率变化,特别是指示表层水骤然降温。北部冬季表层水温从24~25℃降至17~28℃,而南部也从26~27℃降至23~24℃。总的降温趋势与开放西太平洋一致,直接反映了西太平洋暖池在900 ka BP之后MIS22期间有明显的减弱。表层水大幅度降温还发生在后继的MIS 20、18、16几大冰期,说明主要冰期旋回周期由41 ka转变为100 ka经历了长达400 ka的过渡时期,并且冬季风增强也在过渡时期的后半段最明显。南海南北生物组合和δ18O值的差异,突出了中更新世气候转型期边缘海区南北气候梯度反差和冬季风在冰期增强的讯号。结论是：生态环境系统反应总体表现与冰期旋回一致的同时,还包含了独特的地区性系统演变特征。但是,南海—西太平洋地区在0.9 Ma BP前后表层海水盐度因东亚冬季风和海平面下降的定量变化,以及这些变化对气候转型时期海—气耦合过程和生态环境系统的影响,尚缺乏足够的资料和证据。     

格网计算法在空间格局分析中的应用--以贵州景观空间格局分析为例

格网地图与格网计算都是基于空间坐标系统来描述、分析、虚拟区域地理现象的有效方法之一。在空间格局分析应用中具有广阔的前景。景观类型的格网地图不仅能反映它们的空间分布,还能反映它们各自占据的生态空间情况。同时,格网地图还能将景观的空间异质性定量化、空间化和图形可视化表达。在格网地图和格网分析法的基础上,可以利用空间自相关分析、半方差分析等空间统计分析方法对景观的空间格局作进一步的深入研究。     

Intraseasonal oscillation intensity over the western North Pacific:Projected changes under global warming

本文利用8个CMIP5模式的日资料,预估了RCP4.5和RCP8.5情景下全球增温达1.5℃和2.0℃时西北太平洋夏季30～60天和10～20天季节内振荡(ISO)强度的变化情况.大多数模式都认为,无论增温水平或情景如何,预估结果均显示从中南半岛南部到菲律宾以东的带状区域内ISO强度增加,并且关键气象要素背景的变化会对...     

A comparative study between China and U.S. on seismic design philosophy and practice of a long span arch bridge

This paper presents the first of a series of case studies on the seismic design of long span bridges （cable-stayed bridges, suspension bridges and arch bridges） under a cooperative research project on seismic behavior and design of highway bridges between the State Key Laboratory for Disaster Reduction in Civil Engineering, Tongji University and the Multidisciplinary Center for Earthquake Engineering Research, University at Buffalo. The objective of this series of case studies is to examine the differences and similarities on the seismic design practice of long span bridges in China and the U.S., to identify research needs and to develop design guidelines beneficial to bridge engineers in both countries. Unlike short to medium span bridges, long span bridges are not included in most seismic design specifications, mainly because they are location dependent and structurally unique. In this paper, an available model of a steel tied half through arch bridge with a main span of 550m in China is discussed. Analysis is focused on comparisons of the seismic responses due to different ground motions. Seismic design criteria and seismic performance requirements for long span bridges in both countries were first introduced and compared, and then three near field earthquake records with large vertical components were selected as the excitations to examine the seismic behavior and seismic vulnerability of the bridge. Results show that （1） the selected near field ground motions cause larger responses to key components （critical sections） of the bridge （such as arch rib ends） with a maximum increase of more than twice those caused by the site specific ground motions; （2） piers, longitudinal girders and arch crowns are more vulnerable to vertical motions, especially their axial forces; and （3） large vertical components of near field ground motions may not significantly affect the bridge＇s internal forces provided that their peak acceleration spectra ordinates only appear at periods of less than 0.2s. However, they may have more influence on the longitudinal displacements of sliding bearings due to their large displacement spectra ordinates at the fundamental period of the bridge.