时空遥感云计算平台PIE-Engine Studio的研究与应用

随着遥感大数据时代的到来,为快速处理和分析海量遥感数据,国内外涌现了众多遥感云计算平台,使得全球尺度、长时间序列遥感数据的快速分析和应用成为可能。本文在分析国内外遥感云计算平台现状的基础上,针对大数据时代国内缺少功能完备的遥感云计算平台,且国外遥感云计算平台对国产卫星数据支持不足等问题,基于容器云技术,构建了包含国产卫星数据且集数据、算力和技术于一体的时空遥感云计算平台PIE（Pixel Information Expert）-Engine Studio,实现了脚本驱动的遥感数据的按需获取以及海量数据的快速处理。采用Landsat 8数据,以生长季植被指数NDVI（Normalized Difference Vegetation Index）的计算为例,对比了本平台与GEE（Google Earth Engine）的数据处理能力。结果表明,由于计算资源的限制,本平台的计算和导出时间均比GEE稍长,但计算结果的空间分布一致,其中近68%的值均分布在（0.48,0.77）,且二者差值的95.33%集中在（-0.13,0.13）,结果较为可信。因此,本文构建的基于共享、开放的中国自主遥感云计算平台PIE-Engine Studio,可为地球科学领域的研究提供数据和算力支持,将有助于推进中国遥感云计算平台的发展进程,推动国产卫星数据在云计算平台上的应用。     

高放废物地质处置北山预选区旧井和新场花岗岩填隙方解石同位素地球化学

为了探讨高放废物地质处置甘肃北山预选区旧井和新场预选地段的古地下流体来源、成因、化学演化历史以及水-岩相互作用,文章系统研究了该地段花岗岩填隙方解石的产出特征及其碳、氧、锶同位素的组成特征。测试结果表明,旧井和新场方解石的δ13 C组成均较稳定且为负值(分别为-11.6‰~-5.7‰和-9.9‰~-5.1‰);δ18 O分别为-0.7‰~19.7‰和10.9‰~21.9‰,旧井方解石具有更宽的δ18 O取值范围。87 Sr/86 Sr也略有差异,旧井为0.708584~0.718749,新场为0.708838~0.732967,二者随深度增加而呈明显的降低趋势。研究表明,北山预选区地下流体来源及成因较复杂,浅部地下水受大气降水的影响较大,深部流体则主要源于地下咸水,为低温流体蚀变成因。花岗岩裂隙中的水-岩反应强度总体较弱,地下水环境相对稳定。相比较而言,新场岩体深部的地球化学环境更稳定,更有利于高放废物的长期处置。     

Emergency alternative evaluation and selection based on ANP,DEMATEL, and TL-TOPSIS

With the increasing occurrence frequency of emergency events, emergency management (EM) has been a very important issue in management science. One of the major activities of EM is to evaluate and select the most desirable emergency alternative(s). This paper proposes a new framework combining the analytic network process (ANP) method, the decision-making trial and evaluation laboratory (DEMATEL) technique, and 2-tuple linguistic technique for order preference by similarity to an ideal solution (TL-TOPSIS) method to solve the emergency alternative evaluation and selection problem. This study has been done in three stages. In the first stage, we use DEMATEL technique to obtain the network relation map (NRM) among emergency alternative evaluation criteria or sub-criteria. In the second stage, we use ANP method to calculate the global weight of each sub-criterion based on the NRM among emergency alternative evaluation sub-criteria. In the third stage, the ratings of emergency alternative with respect to each sub-criterion are described by linguistic items, and the TL-TOPSIS method is used to rank the emergency alternative. Finally, a practical example of urban fire emergency alternative selection is given to illustrate the application of the proposed framework.

     

崩塌体水力输移与塌岸淤床交互影响试验

以二元结构河岸崩塌模式为研究背景,在弯道水槽中展开系列试验,研究凹岸坡脚处成型崩塌体在水力作用下输移过程及其对岸坡稳定性与河床冲淤的交互影响。试验成果表明,崩塌体经水力分解破碎呈块状、片状或颗粒状起动,部分随水流带至凸岸或下游,堆积在坡脚附近的崩塌体残留量随水力作用大小及土体特性不同而变化。崩塌体临水面周围尤其上下游端水流紊动强烈易形成较大剪切力区,临坡面上下游端附近剪切力较无崩塌体时减小;崩塌体体积越大,对剪切力区特征的改变也越明显。崩塌体的存在虽不能制止附近岸坡的再次崩塌,但可能抑制崩岸发展及附近河床淤积的程度,崩塌体的粘性或体积越大,这种抑制作用越显著;相同崩塌体抑制附近河床淤积的程度较抑制岸坡崩塌的程度大。     

东北地区唐代渤海古城遗址空间格局及影响因素分析

运用ArcGIS软件及空间分析等方法,研究了东北地区234座唐朝渤海国古城遗址分布特征,探讨了古城遗址空间格局与自然条件的关系。通过对234座古城址进行分析,发现唐朝渤海国古城址主要集中在4个集聚区,即图们江?鸭绿江流域集聚区（A）、牡丹江流域集聚区（B）、松花江?辽河流域集聚区（C）、穆棱河流域集聚区（D）。其中图们江?鸭绿江流域集聚区是最主要的遗址分布区,位于研究区的南部,占遗址总量的47.43%。研究区渤海国古代城址呈北疏南密的分布特征,形成明显的“空间组群”格局。在当时的气候环境之下,水热条件等自然因素成为影响城址的主要因素,而政治、经济、军事和交通区位等人文因素则对古城遗址的数量和分布起重要导向作用。     

DOGRAFS温度和风场的预报检验

利用新疆2012年11月1日至2013年10月31日的地面2米温度和10米风场资料,对应用改进后的新疆快速更新循环数值预报同化系统(XJ-RUC)的预报效果进行检验。结果表明：XJ-RUC系统对地面2米温度和10米风场的预报均方根误差和偏差随起报时次和预报时效的不同存在差异,在温度预报方面存在“低温偏高,高温偏低”的趋势,夏、秋、冬三季对18UTC预报最好,春季则对00UTC的预报接近实况;10米风场的预报偏差在冬季随时效推进逐渐增大,春、夏、秋三季则在06UTC出现1.0m/s左右的风速偏差极小值。不论是温度还是风速预报,均在冬季效果最差。     

Empirical formula for wave length of ocean wave in finite depth water

In this paper, function characteristics of dispersion of ocean wave in finite depth water were analyzed systematically. The functional form of the fitting function is reasonably proposed, in which the parameters are optimally determined by the least square method （LSM）. For infinitely deep and extremely shallow water,the fitting function fits strictly the dispersion to be fitted. A new technique is presented in application of LSM.An empirical formula with maximum error of less than 0.5% for computing wavelength in finite depth water is presented for practical applications.     

Effect of growing watershed imperviousness on hydrograph parameters and peak discharge

An increasing impervious area is quickly extending over the Wu‐Tu watershed due to the endless demands of the people. Generally, impervious paving is a major result of urbanization and more recently has had the potential to produce more enormous flood disasters than those of the past. In this study, 40 available rainfall–runoff events were chosen to calibrate the applicable parameters of the models and to determine the relationships between the impervious surfaces and the calibrated parameters. Model inputs came from the outcomes of the block kriging method and the non‐linear programming method. In the optimal process, the shuffled complex evolution method and three criteria were applied to compare the observed and simulated hydrographs. The tendencies of the variations of the parameters with their corresponding imperviousness were established through regression analysis. Ten cases were used to examine the established equations of the parameters and impervious covers. Finally, the design flood routines of various return periods were furnished through use of approaches containing a design storm, block kriging, the SCS model, and a rainfall‐runoff model with established functional relationships. These simulated flood hydrographs were used to compare and understand the past, present, and future hydrological conditions of the watershed studied. In the research results, the time to peak of flood hydrographs for various storms was diminished approximately from 11 h to 6 h in different decrements, whereas peak flow increased respectively from 127 m³ s^?1 to 629 m³ s^?1 for different storm intensities. In addition, this study provides a design diagram for the peak flow ratio to help engineers and designers to construct hydraulic structures efficiently and prevent possible damage to human life and property. Copyright © 2007 John Wiley & Sons, Ltd.     

Semi-empirical models of the white-light flare on October 24, 1991

On October 24, 1991, a white-light flare was observed both from space and from the ground. A multi-waveband spectral analysis shows that the peak time of the continuum emission coincides well with that of a radio burst at 2840 MHz and with the hard X-ray emission. Three semi-empirical models, corresponding to the pre-flare condition and to the peak time of continuum emission both with and without non-thermal excitation and ionization of hydrogen by an electron beam, have been obtained. The results indicate that there is fast heating both in the chromosphere and the photosphere. Some evidence is given that this WLF is very likely a result of bombardment by an electron beam. By taking into account non-thermal effects, the chromospheric temperature of the semi-empirical model is significantly reduced.