高性能计算技术在气象领域的应用

高性能计算通过应用超级计算机与并行处理技术解决复杂的计算问题,是信息技术发展比较迅猛的领域之一。气象应用始终是高性能计算的重要领域,高性能计算技术有效地解决了高分辨率、高精度气象数值预报模式发展限制,在气象预报预测业务中发挥着核心支撑作用。数十年以来,由于数值天气模式研究和业务运行对计算资源的强烈需求,国内外气象领域高性能计算机及应用迅速发展起来。气象领域对高性能计算能力及系统的可靠性需求日益提升。高性能计算技术将与气象预报应用日益融合,相互影响促进,不断创新发展。为满足气象预报预测业务和科研工作需求,中国气象部门将进一步提升高性能计算能力,并致力于优化集约高性能计算系统布局,高效管理计算资源,发挥最大效益。     

辽宁地区不同降水云系雨滴谱参数及其特征量研究

利用位于辽宁省沈阳市和辽阳市的Parsivel（Particle Size and Velocity）激光雨滴谱仪观测到的雨滴谱资料按照积雨云降水系统、积层混合云降水系统和层状云降水系统分析雨滴谱特征量和谱参数的平均特征及随时间的演变特征。结果表明:Gamma分布拟合谱参数N₀和λ按照层状云、积层混合云和积雨云的顺序减小,谱参数μ按照层状云、积层混合云和积雨云的顺序增大;直径小于1 mm的降水粒子对数浓度的贡献最大,直径大于1 mm的降水粒子对雷达反射率的贡献最大;M-P分布的谱参数N₀与雨强I具有幂函数关系,并且随着雨强I的增大而增大,谱参数λ与雨强I具有指数函数关系,随着雨强I的增大而减小。     

Initiation of large-volume silicic centers in the Yellowstone hotspot track: insights from H<Subscript>2</Subscript>O- and F-rich quartz-hosted rhyolitic melt inclusions in the Arbon Valley Tuff of the Snake River Plain

During the onset of caldera cluster volcanism at a new location in the Snake River Plain (SRP), there is an increase in basalt fluxing into the crust and diverse silicic volcanic products are generated. The SRP contains abundant and compositionally diverse hot, dry, and often low-δ¹⁸O silicic volcanic rocks produced through time during the formation of individual caldera clusters, but more H₂O-rich eruptive products are rare. We report analyses of quartz-hosted melt inclusions from pumice clasts from the upper and lower Arbon Valley Tuff (AVT) to gain insight into the initiation of caldera cluster volcanism. The AVT, a voluminous, caldera-forming rhyolite, represents the commencement of volcanism (10.44 Ma) at the Picabo volcanic field of the Yellowstone hotspot track. This is a normal δ¹⁸O rhyolite consisting of early and late erupted members (lower and upper AVT, respectively) with extremely radiogenic Sr isotopes and unradiogenic Nd isotopes, requiring that ~50 % of the mass of these elements is derived from melts of Archean upper crust. Our data reveal distinctive features of the early erupted lower AVT melt including: variable F concentrations up to 1.4 wt%, homogenous and low Cl concentrations (~0.08 wt%), H₂O contents ranging from 2.3 to 6.4 wt%, CO₂ contents ranging from 79 to 410 ppm, and enrichment of incompatible elements compared to the late erupted AVT, subsequent Picabo rhyolites, SRP rhyolites, and melt inclusions from other metaluminous rhyolites (e.g., Bishop Tuff, Mesa Falls Tuff). We couple melt inclusion data with Ti measurements and cathodoluminescence (CL) imaging of the host quartz phenocrysts to elucidate the petrogenetic evolution of the AVT rhyolitic magma. We observe complex and multistage CL zoning patterns, the most critical being multiple truncations indicative of several dissolution–reprecipitation episodes with bright CL cores (higher Ti) and occasional bright CL rims (higher Ti). We interpret the high H₂O, F, F/Cl, and incompatible trace element concentrations in the context of a model involving melting of Archean crust and mixing of the crustal melt with basaltic differentiates, followed by multiple stages of fractional crystallization, remelting, and melt extraction. This multistage process, which we refer to as distillation, is further supported by the complex CL zoning patterns in quartz. We interpret new Δ¹⁸O(Qz-Mt) isotope measurements, demonstrating a 0.4 ‰ or ~180 °C temperature difference, and strong Sr isotopic and chemical differences between the upper and lower AVT to represent two separate eruptions. Similarities between the AVT and the first caldera-forming eruptions of other caldera clusters in the SRP (Yellowstone, Heise and Bruneau Jarbidge) suggest that the more evolved, lower-temperature, more H₂O-rich rhyolites of the SRP are important in the initiation of a caldera cluster during the onset of plume impingement.     

语义感知的地理视频大数据自适应关联组织方法

正地理视频是智慧城市和城市安全领域的重要战略资源。复杂城市环境中公共安全事件大范围流动、多阶段演化及多因素并发等新特征,使现有基于摄像机位置分布存储与分区管理的地理视频组织产生涉案数据检索中迅速扩大的虚警输出规模与急剧降低的涉案信息价值密度。发展支持涉案数据快速整合和案件知识高效理解的数据组织新机制成为发挥地理视频安防价值的重要科学内容。     

采用弯曲进行道路化简冲突避免的方法

针对道路化简后可能导致要素间空间关系前后不一致的问题,提出了一种采用弯曲进行道路化简冲突避免的方法。首先,以弯曲为单元分析道路化简前后弯曲的形态变化规律,对化简过程产生的要素间空间冲突类型进行了归纳和总结;其次,基于道路弯曲与要素之间的空间关系,提出了相应的冲突判别规则;最后,利用弯曲化简的可控性和弯曲组的间接化简策略实现对化简冲突的避免。试验表明,该方法能有效地识别和避免道路化简产生的空间冲突,确保化简前后道路与其他要素空间关系的一致性。     

顾及邻域居民地群组相似性的道路网匹配方法

现有多源同比例尺道路网匹配方法中,大多只利用道路自身特征进行匹配,而较少顾及道路周边要素对匹配过程的影响和约束,从而影响了道路网匹配效果的进一步提高,特别是对系统误差改正后仍存在一定位置或旋转偏差的道路数据进行匹配时,这种影响尤为明显。本文借鉴人类对陌生环境的空间认知特点,提出了一种顾及邻域居民地群组相似性的道路网匹配方法。该方法通过构建城市骨架线网确定与道路相邻的居民地群组,进而计算居民地群组空间关系和几何特征相似度来获得对应道路的匹配结果。其特点在于:对存在位置或旋转偏差的道路数据匹配,以其邻域空间内居民地群组的整体相似性指标来带动道路自身匹配,实际上是增加了周边居民地群组对道路匹配过程的约束,更具鲁棒性。试验及对比分析表明,本方法能够较好地解决系统误差改正后仍存在较大位置和旋转偏差的道路数据间的匹配问题,提高匹配的正确率。     

全球气候数据集生成及气候变化应用研究

科技部在"十三五"期间部署的国家重点研发计划"全球变化及应对"专项资助了"全球气候数据集生成及气候变化关键过程和要素监测"研究项目。项目围绕由全球气候观测系统提出的基本气候变量,完善地空天基观测体系,生成中国首套以遥感数据为主体的涵盖大气、海洋和陆表长时间序列、高精度、高时空一致性的产品,即气候数据集,动态监测全球变化关键过程和要素。     

基于 Lidar 数据和倾斜摄影的城市三维模型构建

针对传统建模方式建模效率低、模型纹理不够丰富真实等缺点,阐述一种以机载Lidar技术来获取地面点高精度的三维坐标、用倾斜摄影测量为三维建模提供纹理的一种新兴的三维建模方式。实验表明,该建模方式不仅工作效率高、真实反应城市空间布局,而且定位精度高、具备可量测分析功能等,同时提供DOM,DEM,点云数据,倾斜影像等数据。     

Spatial representation in the social interaction potential metric: an analysis of scale and parameter sensitivity

The social interaction potential (SIP) metric measures urban structural constraints on social interaction opportunities of a metropolitan region based on the time geographic concept of joint accessibility. Previous implementations of the metric used an interaction surface based on census tracts and the locations of their centroids. This has been shown to be a shortcoming, as the metric strongly depends on the scale of the zoning system in the region, making it difficult to compare the SIP metric between metropolitan regions. This research explores the role of spatial representation in the SIP metric and identifies a suitable grid-based representation that allows for comparison between regions while retaining cost-effectiveness with respect to computational burden. We also report on findings from an extensive sensitivity analysis investigating the SIP metric’s input parameters such as a travel flow congestion factor and the length of the allowable time budget for social activities. The results provide new insights on the role of the modifiable areal unit problem in the computation of time geographic measures of accessibility.     

Recursive algorithm for fast GNSS orbit fitting

Gaussian elimination is an efficient and numerically stable algorithm for estimating parameters and their precision. However, before estimating the parameters, it is often prudent to perform statistical tests to achieve the best fitting model. We use Gaussian elimination to select the best fitting model among candidate models. A succinct relationship between the weighted sum of squared residuals and the previous one is revealed by a volume formula. For quick parameter estimation and determination of weighted sum of squared residuals, a recursive elimination algorithm is proposed in the context of Gaussian elimination. In order to improve the model selection efficiency, the parameter estimation and the determination of the weighted sum of squared residuals are carried out in parallel using the proposed recursive elimination algorithm in which the improvement at each recursive stage is judged by the Bayesian information criterion. Ultimately, the computational complexity and numerical stability of the recursive elimination proposed are briefly discussed, and a GNSS orbit interpolation example is used to verify the results. It shows that the proposed recursive elimination algorithm inherits the numerical stability of the Gaussian elimination, and this algorithm can be used to examine the gain from the newly introduced parameter, dynamically assess the fitting model, and fix the optimal model efficiently. The optimal fitting model with the lowest information is very close to the real situation verified by checkpoints.