The Zeeman effect: applications to solar physics

This contribution is intended to give a brief review of some of the results concerning the Zeeman effect which have been recently published in the literature or which appear to be particularly relevant at the light of recent technological improvements in observations. The arguments emphasized are the Zeeman effect in molecular lines, the asymmetries observed in Stokes profiles from sunspots, and the interpretation of spectropolarimetric observations in the infrared.     

基于GPU混合反演的隧道电阻率超前探测成像研究

隧道施工期超前探测对于避免突涌水灾害的发生具有重要作用,为满足隧道三维电阻率超前探测快速化解译与成像的要求,本文提出了一种基于GPU并行的蚁群算法与最小二乘方法相结合的混合反演算法.该方法结合线性反演与非线性反演的优点,利用蚁群算法全局搜索能力强的优点为最小二乘反演提供较优的初始模型,以克服最小二乘算法容易陷入局部最优的缺点,提高了隧道三维电阻率反演成像的精度.同时,基于蚁群算法的天然并行性,提出了CUDA环境下的GPU并行策略,实现了三维电阻率反演的快速化成像.其次,开展了基于GPU混合反演的数值算例,与传统最小二乘线性反演进行了对比,基于GPU并行计算的混合反演计算效率得到了显著提高,对含水构造的位置、形态有较好的反映,压制了三维反演的多解性.最后开展了物理模型试验,结果表明基于GPU混合反演探测的低阻异常体与实际含水构造的位置较为相符,发现基于GPU加速的混合反演方法在提高探测精度与加快反演速度方面具有显著优势,为三维电阻率混合反演方法在隧道超前探测实际工程中的应用奠定了基础.     

引黄水源水库平-丰水期有色可溶性有机物(CDOM)的来源、组成及差异分析

为探究引黄水源水库——门楼水库平水期和丰水期有色可溶性有机物(CDOM)的组成特征、来源及差异,运用紫外—可见光谱技术(UV-vis)和三维荧光光谱(EEMs)技术,结合平行因子分析法(PARAFAC)分析2022年5月(平水期)和2022年7月(丰水期)有色可溶性有机物含量及组分变化。研究结果表明:PARAFAC识别出2类荧光组分,分别是C1(Ex=355 nm,Em=476 nm,类腐殖质组分)和组分C2(Ex=225 nm,Em=320 nm,类蛋白组分);丰水期CDOM组分荧光强度显著高于平水期。CDOM光谱参数表明,门楼水库水体处于中营养状态,水体CDOM受新生内源和外源输入共同影响,以自生源为主;水库CDOM具有相对分子量小、腐殖化程度较弱的特点。丰水期水库富营养化水平和CDOM相对浓度低于平水期;丰水期CDOM疏水性组分比例和芳香化程度高于平水期。水质理化指标、CDOM组分和光谱参数相关性分析结果表明SUVA₂₆₀和SUVA₂₈₀与DOC呈显著负相关,说明紫外—可见光谱参数在一定程度上可以用来估算DOC的浓度;Chl.a浓度作为...     

A SUPER-DYNAMIC DEFORMATION MEASUREMENT SYSTEM WITH LONG-TIME PARALLEL CONTINUOUS ACQUISITION

In the simulation experiments of earthquakes in laboratory, the instability slip or rupture events are obtained through steady state loading to simulate earthquake processes. In the experiments, steady-state deformation and unstable sliding occur alternately. It is hard to determine the origin time and duration of the instability event of fault, and there may be many instability events in one experiment. Therefore, in order to ensure that sufficient data is obtained at the extremely short instability moment to analyze the mechanical process of the earthquake source, the data acquisition system is required to continuously collect data at high-speed from the beginning of the experiment until the end, and the lasting time can be more than ten hours, so it requires huge storage space. Although the upper frequency limit of the instability signal is unknown exactly, but the previous experiments have shown that the frequency of the signal will reach hundred to several kilohertz, so the sampling frequency of the data acquisition should be above megahertz. In this case of long-time continuous high-frequency data acquisition, it is still necessary to maintain a high signal-to-noise ratio. Furthermore, previous studies have proved that the source mechanics field has a complex spatial structure, which is difficult to describe with a few measuring points, and it is necessary to perform simultaneous measurements of the source mechanics field for dozens or more measuring points. The combination of long-term continuous recording, high-resolution high-frequency sampling and multi-point simultaneous measurement poses a huge challenge to the technical indicators of the observing system. With the method for composing distributed synchronous acquisition machine group by using multiple high-resolution high-frequency sampling computers, a super dynamic deformation measurement system of high signal-to-noise ratio, which features 64-channel, 16-bit resolution, 4MHz sampling frequency, and parallel continuous acquisition with tens of hours was developed. This system can realize the synchronous acquisition of various signals, such as strain, acoustic emission, electromagnetic waves and displacement, so it is convenient for analyzing the conversion relationship between various physical quantities.     

Fast Geographically Weighted Regression (FastGWR): a scalable algorithm to investigate spatial process heterogeneity in millions of observations

Geographically Weighted Regression (GWR) is a widely used tool for exploring spatial heterogeneity of processes over geographic space. GWR computes location-specific parameter estimates, which makes its calibration process computationally intensive. The maximum number of data points that can be handled by current open-source GWR software is approximately 15,000 observations on a standard desktop. In the era of big data, this places a severe limitation on the use of GWR. To overcome this limitation, we propose a highly scalable, open-source FastGWR implementation based on Python and the Message Passing Interface (MPI) that scales to the order of millions of observations. FastGWR optimizes memory usage along with parallelization to boost performance significantly. To illustrate the performance of FastGWR, a hedonic house price model is calibrated on approximately 1.3 million single-family residential properties from a Zillow dataset for the city of Los Angeles, which is the first effort to apply GWR to a dataset of this size. The results show that FastGWR scales linearly as the number of cores within the High-Performance Computing (HPC) environment increases. It also outperforms currently available open-sourced GWR software packages with drastic speed reductions – up to thousands of times faster – on a standard desktop.     

北京市公共服务设施集聚中心识别分析

公共服务设施集聚中心不仅是多样化公共服务设施分布的集中区,也是公共服务设施消费的热点区和城市活力的窗口区。已有城市多中心研究主要关注就业和人口多中心,却较少关注公共服务设施多中心特征。基于北京市公共服务设施空间点要素数据,结合不同类型和等级公共服务设施的服务半径与质量特征,采用加权核密度与等值线分析等方法对北京市公共服务设施集聚中心进行了识别。研究发现：不同类型公共服务设施热点区分布的空间模式有所差异,但均存在一定程度的中心边缘结构;公共服务设施集聚强度和混合度分别呈现出“一心五片”与“一核多点”的空间特征;根据公共服务设施集聚强度和混合度的等值线分布综合判定,研究区范围共识别出136个公共服务设施集聚中心,其集聚强度与距市中心距离呈现出U型变化规律,与所在街道的人口密度存在指数分布规律;多元回归模型验证,最近公共服务设施集聚中心距离对北京城市居民公共服务设施满意度具有显著的负向影响,且其相对影响强度超过“市中心距离”区位变量;北京市公共服务设施集聚中心的形成机制包括自然历史因素的基础作用、经济发展因素的主导作用、社会需求因素的调节作用和规划政策因素的引导作用。研究认为应加强北京城市公共服务设施集聚中心的空间均衡化建设,适度增加城市边缘郊区公共服务设施集聚中心的数量和服务能级,有助于引导和疏解非首都功能产业和人口向城市郊区转移。     

面向大数据的时空信息云平台建设研究

在大数据时代来临的背景下,传统数据处理和分析方法已无法满足时空大数据高效存取、实时处理、分析挖掘等需求。本文结合智慧广东时空信息云平台的项目建设案例,探讨了面向大数据的时空信息云平台的建设目标、总体架构、关键技术、应用情况等内容,为解决海量时空大数据处理与分析性能问题,推动地理信息服务高质量发展做了积极探索。     

大规模数据集Spark并行优化谱聚类

针对已有大规模数据集并行谱聚类算法的计算耗时和资源占用巨大等问题,基于当前批处理和图计算兼顾的Spark并行技术,提出了大规模数据集谱聚类的并行优化改进算法,算法通过并行单向迭代避免了相似矩阵计算时的数据重复计算,通过并行位置变换、标量乘法替换及距离缩放优化算法的资源占用,通过近似特征向量替代进一步优化算法的计算量。试验结果验证了算法近特征向量的有效性及在大规模数据集下良好聚类性能和扩展性。     

自适应分块加权Wallis并行匀色

针对区域范围内多幅待镶嵌影像之间的色彩差异问题,提出一种基于GPU的分块加权Wallis并行匀色算法。首先,根据变异系数对影像自适应分块并利用双线性插值确定每一个像素的变换参数,利用加权Wallis变换消除影像间的色彩差异。然后,为了控制区域整体的匀色质量,利用Voronoi图和Dijkstra算法确定影像间的处理顺序。最后,利用GPU技术进行并行任务设计并从配置划分、存储器访问和指令吞吐量等方面进行优化,提高算法运算效率。实验结果表明,本文方法既能有效地消除影像间色彩差异,又能消除影像间的对比度差异。与CPU串行算法相比,GPU并行算法显著减少了计算时间,加速比最高达到60倍以上。     

Possibilities of investigation of interstellar matter by radio recombination lines (Review)

Radio recombination lines (RRL) discovered about 25 years ago became a very effective tool for the investigation of interstellar matter (ISM). These lines are unique by the number of transitions and the spectral range observed: from 2 mm to 20 m wavelength. This allows to study the ISM under variety of physical conditions which differ by 7 to 8 orders of magnitude in density and 3 orders of magnitude in temperature. RRL allow to determine the main physical conditions in HII and HI-CII regions, to map the galactic distribution of ionized hydrogen, and to obtain the helium abundance of the ISM. Furthermore, the information contained in the RRL allows to estimate the intensity of cosmic rays and to draw conclusions on the galactic evolution. The physical properties of the RRL and the results obtained by observations of RRLs are reviewed.