康乐气象站迁移前后地温同期观测资料对比分析及订正

为了消除城市化进程对旧站地温的影响,运用数理统计方法对康乐县新旧地面气象观测站2012年1-12月同期0-20 cm地温平均值、方差及相关性进行F、T、t检验,对新旧2站数据差异显著的地温,利用最小二乘法进行最佳线性拟合,建立2站地温订正方程。结果表明：整体而言,新站与旧站0-20 cm地温均存在负温差,旧站地温较新站的偏高,大多数月份差值较小,资料连续性较好,但2站地温差值变幅较大。地温差异显著的1月和10月2站地温资料呈显著线性相关,经订正后的新旧站资料基本重合。     

浅基础地基荷载-变形特性的分析研究

在土体主应力方向的旋转过程中,主应力方向与塑性主应变增量方向之间存在着非共轴现象,非共轴现象对应力-应变关系具有显著的影响。通过有限元二次开发,将角点结构非共轴理论应用到有限元程序ABAQUS中;对密砂单剪试验进行数值模拟,研究了非共轴现象及其对应力-应变关系的影响;采用离心模型试验方法,对圆形浅基础作用下饱和密砂地基荷载-变形特性进行试验研究;对离心模型试验进行数值模拟,将试验结果与数值计算结果进行对比,对非共轴模型的计算结果的合理性进行验证。研究结果表明:非共轴现象对浅基础地基荷载-变形特性具有显著的影响;当选取合理的非共轴塑性模量时,非共轴模型的计算结果与离心模型试验结果比较接近。     

海洋重力仪稳定性测试与零点漂移问题

稳定性指标是海洋重力仪最重要的技术特性之一。针对国内对海洋重力仪稳定性测试与评估重视不够、数据处理过程欠规范、技术指标要求欠细化等现实问题,研究探讨了海洋重力仪稳定性测评的技术流程和数据处理方法,重点分析了环境因素和重力固体潮效应对测试结果的影响,提出了重力仪零点趋势性漂移、有色观测噪声与随机误差的分离方法,建立了比较完善的海洋重力仪稳定性评估指标体系,分析论证并进一步明确了重力仪零点漂移非线性变化的限定指标要求,为修订现行海洋重力测量作业标准提供了可靠的理论依据。     

《海道测量规范》修订的若干问题

根据15年来GB12327-1998《海道测量规范》的使用情况和海洋测绘新技术的发展,针对《海道测量规范》的指导思想、框架结构、技术方法、技术指标等方面进行了探讨,提出了相应的修订建议,以进一步增强规范的先进性、科学性、权威性。     

机载激光测深系统研制中的关键技术

根据机载激光测深系统的工作机理,分析梳理了研制过程中涉及的主要关键技术,包括:高重复频率激光器研制、大动态范围微弱光信号提取、海底回波识别、海面波浪改正、浅水与深水分离测量、高速多通道数字采集、海陆分界识别、定位测姿、水位改正和测深数据质量控制等技术。只有成功突破上述关键技术,才能实现《海道测量规范》要求的测深精度指标。     

一种语音识别的可定制云计算方法

以云的方式提供语音识别服务是1种新的趋势。当前市场上的语音云都是面向通用领域的,针对这一现状,提出了1种面向特定领域,可定制的,可满足用户个性化需求的云计算方案。首先介绍了语音识别的MapReduce过程,接着描述了基于Sector/Sphere的语音识别系统组成及语音识别的Sphere处理流程,并给出了语音识别系统的云计算架构,然后探讨了语音识别的可定制实现,最后进行实验评估,证明云计算架构的语音识别系统与传统C/S架构的语音识别系统相比更具优势,且具有一定的可扩展性。     

太平洋鳕鱼脑中硫苷脂的分离纯化和分析

以太平洋鳕鱼为原料,从鳕鱼脑中分离硫苷脂,确立了提取及纯化的条件,并对硫苷脂的纯度和分子种进行分析。首先,采用氯仿/甲醇(2∶1,v/v)提取总脂。然后,依次采用氯仿/甲醇/水(7∶3∶0.3,v/v/v)洗脱硅胶柱层析,含0.2mol/L乙酸铵的氯仿/甲醇/水(30∶60∶8,v/v/v)洗脱DEAE Sephadex-A25离子交换柱层析,40%甲醇脱盐和100%甲醇洗脱反相C8柱层析,获得硫苷脂纯品。最后,利用500YMC Diol液相色谱柱(3.0mm×250mm,5μm)分离,以正己烷/异丙醇(70∶30,v/v),异丙醇/水/甲酸/氨水(100∶13∶1∶0.14,v/v/v/v)为流动相梯度洗脱,采用负电喷雾电离(ESI)和母离子扫描模式,用液相色谱-串联质谱(LC-MS/MS)法分析了鳕鱼脑硫苷脂的纯度和分子种组成,并比较了其与哺乳动物脑中硫苷脂分子种组成的异同。结果表明,本实验制得的鳕鱼脑硫苷脂纯度为90.74%,与哺乳动物类似,鳕鱼脑硫苷脂的长链碱基以鞘氨醇为主,主要分子种为d18∶1-C24∶1,但脂肪酸的羟基化程度略低,且含少量独特的分子种,如d18∶1-22∶1和d18∶2-25∶2。     

Optimal Forward-Scattering Angles of Atmospheric Aerosols in North China

The accurate understanding of atmospheric aerosol extinction coefficients is very important for at- mospheric science research. To achieve a fast and simple method for determining the parameters, the selection of optimal forward-scattering angles of atmospheric aerosols is required. In this paper, the authors introduce the detec- tion basis of forward-scattering of atmospheric aerosols, and the authors verify the sensitivity of the phase function to the real part of the complex refractive index. The au- thors use the Jaenicke urban aerosol model to determine that forward-scattering angles near 33° are suitable. However, the optimal forward-scattering angles in North China are between 37° and 40°. Numerical simulation shows that certain types of particle size distribution of newly generated particles and pollution have limited in- fluences on the selection of forward-scattering angles. But the ranges of these insensitive angles shift - 10 degrees for dust intrusion, and the relative deviations of the phase function are less than 5.0% within extra angles of 0° to 3°. This study can serve as a reference for the selection of optimal forward-scattering angles for visibility meters and Present Weather Identifiers （PWIs） in addition to the de- tection of forward-scattering optical properties.     

Assimilating the LAI Data to the VEGAS Model Using the Local Ensemble Transform Kalman Filter: An Observing System Simulation Experiment

Information on the spatial and temporal pat- terns of surface carbon flux is crucial to understanding of source/sink mechanisms and projection of future atmospheric CO2 concentrations and climate. This study presents the construction and implementation of a terrestrial carbon cycle data assimilation system based on a dynamic vegetation and terrestrial carbon model Vegetation-Global-Atmosphere-Soil （VEGAS） with an advanced assimilation algorithm, the local ensemble transform Kalman filter （LETKF, hereafter LETKF-VEGAS）. An observing system simulation experiment （OSSE） framework was designed to evaluate the reliability of this system, and numerical experiments conducted by the OSSE using leaf area index （LAI） observations suggest that the LETKF -VEGAS can improve the estimations of leaf carbon pool and LAI significantly, with reduced root mean square errors and increased correlation coefficients with true values, as compared to a control run without assimilation. Furthermore, the LETKF-VEGAS has the potential to provide more accurate estimations of the net primary productivity （NPP） and carbon flux to atmosphere （CFta）.     

Phylogenomic analysis of transcriptomic sequences of mitochondria and chloroplasts for marine red algae (Rhodophyta) in China

The chloroplast and mitochondrion of red algae （Phylum Rhodophyta） may have originated from different endosymbiosis. In this study, we carried out phylogenomic analysis to distinguish their evolutionary lin-eages by using red algal RNA-seq datasets of the 1 000 Plants （1KP） Project and publicly available complete genomes of mitochondria and chloroplasts of Rhodophyta. We have found that red algae were divided into three clades of orders, Florideophyceae, Bangiophyceae and Cyanidiophyceae. Taxonomy resolution for Class Florideophyceae showed that Order Gigartinales was close to Order Halymeniales, while Order Graci-lariales was in a clade of Order Ceramials. We confirmed Prionitis divaricata （Family Halymeniaceae） was closely related to the clade of Order Gracilariales, rather than to genus Grateloupia of Order Halymeniales as reported before. Furthermore, we found both mitochondrial and chloroplastic genes in Rhodophyta under negative selection （Ka/Ks〈1）, suggesting that red algae, as one primitive group of eukaryotic algae, might share joint evolutionary history with these two organelles for a long time, although we identified some dif-ferences in their phylogenetic trees. Our analysis provided the basic phylogenetic relationships of red algae, and demonstrated their potential ability to study endosymbiotic events.