水利大数据研究综述          下载免费PDF全文

陈军飞  邓梦华  王慧敏 《水科学进展》2017,28(4):622-631

为了推动水利大数据的发展及其在水利管理中的应用,对水利大数据相关的研究进行总结和分析,并对未来的研究方向进行展望。在回顾水利大数据的产生背景和发展历程的基础上,界定了水利大数据的概念及内涵;构建了水利大数据的研究框架,并对水利大数据的存储与共享方法、分析平台、挖掘算法、数据可视化和应用等进行了探讨。结合水利大数据研究发展和应用需求,从数据集成方法、分析挖掘算法、数据安全和综合决策分析平台等方面对水利大数据研究领域的发展趋势进行阐述。  相似文献   

Elements’ Geochemical Characteristics of PM10 and PM2.5 in Beijing During Winter and Spring     

Wang Di  Feng Haiyan  Jing Huimin 《地球科学进展》2017,32(8):850-858

From 2012 to 2013, heavy haze frequently hit Beijing in spring and winter. The fine atmospheric particulates can be inhaled by people, and remain in the respiratory tract and lung for quite a long time. The heavy metal elements in the particles are harmful, and even carcinogenic to human bodies. Therefore, it is necessary to master the geochemical characteristics and the temporal and spatial distribution of the heavy metal elements in atmospheric particles. The atmospheric particulates (TSP/PM₁₀/PM_2.5) were collected by using TH1500C intelligent medium volume (80~120 L/min) air samplers in the five functional areas and suburbs of Beijing, respectively in January 2013 (heating period) and April 2013 (non-heating period). The five functional areas were: building materials factory area, residential area, education area, business area and recreation area, each functional area having three sampling sites, and five in suburbs. The sampling height was 1.5 m above the ground and the distance of the sampling sites to roads exceeded 50 m so as to avoid excessive impact of vehicle exhaust emissions. These samples were analyzed by Inductively Coupled Plasma-Mass Spectrometry (ICP-MS), Inductively Coupled Plasma Optical Emission Spectrometer(ICP-OES)and Atomic Fluorescence Spectrometer (AFS), by which 21 elements including Al, Fe, Mn, As, Hg, Cd, Cr were tested. Based on the comparisons of the concentration and element content variation of atmospheric particulates of these functional areas in and after the heating period, the spatial distribution of atmospheric particulates and the influence of coal combustion on the concentration and composition of atmospheric particulates were revealed. The elements sources of atmospheric particulates were also discussed by computing the enrichment factor of the elements, providing a scientific basis for the air contaminant treatment in Beijing. The results showed that the total concentration of the 21 elements of PM₁₀ and PM_2.5 in the functional areas of Beijing in winter was higher than that in spring, the most marked among them being the business area. In winter, the concentration of pollution elements in the building materials factory area and the business area in Beijing was extremely high, and the residential area, education area, recreation area and suburbs where people live were much better, among which the education area was the best. The concentration change of particulates in suburbs was quite smaller in winter and spring than that of the urban area. It also showed that the concentrations of Cd, Cr, As and Hg in PM₁₀ increased by 233%,306%,298%,141%,respectively and the increase in PM_2.5 was 442%、309%、310%、256%, respectively. These abovementioned elements show a tendency to accumulate mainly in PM_2.5 whose concentrations in winter were significantly higher than those in spring. It indicates that coal heating during winter makes great contributions to the polluting elements in atmosphere and the main polluting particulates are Cd, As and Hg.  相似文献   

黑龙江三道湾子金矿床火山岩-次火山岩年代学、地球化学和地质意义     

程琳  彭晓蕾  韩吉龙  任亮  刘阳 《世界地质》2017,36(2):460-473

三道湾子金矿床位于小兴安岭北麓,是典型的以碲金化合物为主的矿床,为了限定其成矿时代和厘定其成矿地质背景,笔者对同成矿地质体火山岩和成矿期后的玄武粗安玢岩的单颗粒锆石U-Pb年龄和元素地球化学进行测定,获得同成矿地质体粗面英安斑岩及玄武粗安玢岩的~(206)Pb/~(238)U年龄加权平均值分别为(124.2±0.9)Ma(MSWD=0.5,n=18)和(118.9±2.3)Ma(MSWD=0.5,n=6);赋矿围岩为玄武粗安岩与同成矿地质体粗面英安斑岩,Si O2含量为52.75%~62.30%,K_2O和Na_2O含量分别为2.02%~2.85%和3.96%~4.24%。里特曼指数(δ)在0.85~3.67之间,属于钙碱性到高钾钙碱性岩系特征;成矿期后脉岩玄武粗安玢岩,SiO_2含量为51.53%~52.32%,K_2O含量为2.30%~2.36%,Na_2O含量为3.78%~4.14%,里特曼指数(δ)在4.05~4.86之间;微量和稀土元素均显示富集大离子亲石元素(Cs、K、Rb、Ba、LREE等),相对亏损高场强元素(Nb、Ta、Zr、Hf等)和HREE,特别是强烈亏损Th、U和Nb、Ta的特征,与岛弧钙碱性玄武岩相似。研究揭示了该矿床形成于早白垩世(124~119 Ma)或晚中生代造山后伸展构造环境,成矿热动源可能是交代大洋性质的岩石圈地幔。  相似文献   

云南先峰中新世木乃伊化木化石α-纤维素的提取     

任军博  杨思宇  苏涛  臧利斌  全成 《世界地质》2017,36(3):1001-1007

本文对云南先锋中新世木乃伊化木化石进行α-纤维素实验提取,以验证化石纤维素的可得性。实验利用有机溶剂分离、碱液分离、无机酸分离等方法分别对研究区5件木乃伊化化石样品进行处理,并采用红外光谱检测方法与现代木材α-纤维素和国际标准α-纤维素进行对比试验。实验结果显示,3种提取方法均可以从化石木材中提取出α-纤维素,但过程特征各有不同。其中,有机溶剂分离实验过程耗时最长,碱液分离方法样品损失最大。相比之下,无机酸分离方法对于木乃伊化木化石α-纤维素提取的效率最高。  相似文献   

基于众源轨迹数据的人造地表覆盖增量信息提取     

高永伟  周晓光  任常青 《地理信息世界》2017,24(3)

目前,我国已研制出2000年、2010年两期30m分辨率全球地表覆盖数据集。为了保持该数据集的现势性,需要及时发现全球地表覆盖增量信息并更新相应数据。采用遥感影像、测绘数据提取全球地表覆盖变化信息的传统方式,存在费用高,不及时等问题。近年来众源地理数据迅猛发展,特别是由于众源轨迹数据具有免费共享、可靠性强与准实时等优点被广泛应用在交通道路信息提取的应用中。经分析我们发现轨迹数据与人造地表覆盖之间存在一定的对应关系,因此,本文研究提出了一种采用众源轨迹提取人造地表覆盖增量信息的方法。该方法首先对原始轨迹数据进行预处理,然后将轨迹数据栅格化,再依据轨迹点栅格覆盖比率进行人造地表覆盖数据粗提取并去除微小孔洞及独立栅格单元,最后将其与全球人造地表覆盖数据作栅格运算提取人造地表覆盖增量信息,并用纽约市实际轨迹数据验证了该方法的可行性。  相似文献   

在线地图集发布与服务系统的设计与实现     

王琨  周启  任福  任东宇  杜清运  鲁月新 《地理信息世界》2017,24(1)

地图集作为反映一个地区或一种领域研究成果的地图集合,是测绘行业的一个重要组成部分,传统地图集编制与发布的成果多为纸质版,制作与发布周期比较长,更新速度慢。该系统对在线地图集发布与服务系统进行开发,有效地缩短了地图集编制时间,提高了地图集编制与发布的效率。在对整个系统进行了完整的设计后,对相关关键技术进行了研究,最后实现了整个系统的开发,并对系统主要功能做出展示。  相似文献   

基于OSM的地表覆盖变化数据提取     

李志盛  周晓光  任常青 《地理信息世界》2017,24(2)

2014年,中国研制出了2000年和2010年两期30 m分辨率的全球地表覆盖数据产品,为了保持该数据集的现势性,持续更新成为该领域关键问题。变化数据的快速收集是全球地表覆盖数据更新的核心,志愿者地理信息(volunteered geographic information,VGI)的兴起为解决此问题提供了一条新的途径。在目前VGI项目中,OpenStreetMap(OSM)具有数据丰富、现势性强等优点,可以作为全球地表覆盖数据更新的参考信息来源,然而OSM数据不能直接对应地表覆盖增量数据,因此,本文提出了一种基于OSM的地表覆盖增量数据提取方法。该方法首先以OSM定义的几何类型和属性数据作为分类依据,建立OSM数据到地表覆盖数据的转换规则;然后针对类型转换后数据存在的目标零散等问题,采用四叉树聚类方法对转换数据进行综合;继而对整合后的数据栅格化并与GlobeLand30数据进行叠加运算得到地表覆盖增量数据。最后以爱沙尼亚区域的OSM数据验证了本文方法的可行性。  相似文献   

Ore-forming fluid characteristics and genesis of vein-type lead-zinc mineralization of Xiaohongshilazi deposit,Jilin Province,China     

LI Yong  REN Yunsheng  HAO Yujie  YANG Qun 《世界地质(英文版)》2017,20(4)

The Xiaohongshilazi mineral deposit in Jilin Province,China,is located in the accretion zone in the northern margin of the North China Block. The deposit contains two types of ore bodies: layered Pb-Zn ore bodies in volcanic rock and vein-hosted Pb-Zn ore bodies controlled by fractures. The vein Pb-Zn ore bodies are strictly controlled by tectonic fracture zones trending in S-N direction,which comprise sulfide veins or sulfidebearing quartz veins distributed along faults or structural fissures. The ores mainly appear mesh-vein and vein structures,and also show solid-solution separation and metasomatic textures. The metal minerals are mainly sphalerite,galena,and pyrite,etc. Wall-rock alteration includes mainly sericitization,chloritization,silicification and carbonatization,etc. Microscope observations and Raman spectroscopy analyses indicate that the oreforming fluid of the vein Pb-Zn ore bodies was mainly magmatic water with low temperature,low salinity,and a shallow depth of metallogenesis( ~ 1.5 km). Sulfur and lead isotope analyses indicate that the sulfide source is mainly formation sulfur or biogenic sulfur,which is similar to the sulfur source of hydrothermal deposit( negative( δ~(34) S values),while the main Pb source was the upper crust with some mantle input. This article argues that the vein Pb-Zn ore body of the Xiaohongshilazi deposit is a low-to medium-temperature hydrothermal vein type related to the formation of a shallow magma chamber.  相似文献   

基于贝叶斯机器学习的生态模型参数优化方法研究     

何立杰  何洪林  任小丽  葛蓉  杨涛  朱超 《地球信息科学学报》2017,19(10):1270-1278

参数优化方法是准确估计生态模型参数、降低其不确定性的有效手段。本文提出一种基于贝叶斯机器学习的No-U-Turn Sampler（NUTS）生态模型参数优化方法。NUTS是一种高效的参数优化方法,每次取样中利用递归算法生成候选参数集（二叉树）推断参数的后验信息,如果满足约束条件“非U型回转”,不断构建子树更新参数;否则,记录本次抽样的“最优”参数集,并开始下一次取样,直到获取足够样本。该算法在每次取样中充分优化参数,避免因随机游走行为产生冗余抽样,提高了参数优化效率。本文以千烟洲亚热带人工针叶林碳通量模拟为例,基于Pymc3框架利用NUTS参数优化方法实现了碳通量（Net Ecosystem Exchange,NEE）模型参数反演,并与Metropolis-Hastings（MH）方法进行对比。结果表明,本文算法的参数值达到稳定波动时的抽样次数减少了85%左右,参数优化效率提升3倍左右。参数优化后,2种NEE模型中7个参数不确定性降低10%~53%。此外,NEE模拟效果明显提升,模拟值与实测值的R²分别提高23%和17%,RMSE分别降低3%和4%。综上所述,本文提出的参数优化方法对生态领域的参数估计或数据同化工作具有一定的借鉴意义。  相似文献   

面向户外增强现实的地理实体目标检测     

乔延军  饶锦蒙  王骏星  杜清运  任福 《地理信息世界》2017,24(5)

本文研究了多种计算机视觉领域中用于图像分类和目标检测的卷积神经网络模型,在一种目标检测模型的基础上进行了修改,并结合了残差神经网络结构,旨在解决户外增强现实中地理实体目标检测的问题。我们根据这一模型开发了一款户外增强现实原型系统,通过将训练好的模型作为目标检测引擎,达到了较好的近实时性和准确性。  相似文献