A comparison of three marine ecosystems surrounding the Korean peninsula: Responses to climate change

This study uses a comparative approach to examine responses of marine ecosystems to climatic regime shifts. The three seas surrounding the Korean peninsula, the Japan/East Sea, the East China Sea and the Yellow Sea represent three contiguous but distinct ecosystems. Sampling has been carried out by the National Fisheries Research and Development Institute of South Korea since 1965, using the same methods in all three seas. Sampling was generally synoptic. Amplitude time series of 1st EOF modes for temperature, salinity, zooplankton biomass and concentrations of four major zooplankton taxa were used to determine whether the three marine ecosystems respond in a similar manner to climate variations. Temporal patterns of the variables were strongly similar among the three seas at decadal time scales, but very weakly similar at interannual scales. All three seas responded to a climatic regime shift that occurred in 1989. Temperature, zooplankton biomass and copepod concentrations increased in the late 1980s or early 1990s in all three seas. Concentrations of amphipods, chaetognaths and euphausiids also increased in the Japan/East Sea and the East China Sea, but not the Yellow Sea. The Yellow Sea ecosystem differs strongly from the other two seas, and water exchange between the Yellow Sea and the East China Sea is much weaker than that between the East China Sea and Japan/East Sea. Spatial patterns of zooplankton determined by the EOF analysis were closely related to currents and fronts in each of the three seas.     

Numerical Simulation of Non-Linear Wave Propagation in Waters of Mildly Varying Topography with Complicated Boundary

In this paper, the characteristics of different forms of mild slope equations for non-linear wave are analyzed, and new non-linear theoretic models for wave propagation are presented, with non-linear terms added to the mild slope equations for non-stationary linear waves and dissipative effects considered. Numerical simulation models are developed of non-linear wave propagation for waters of mildly varying topography with complicated boundary, and the effects are studied of different non-linear corrections on calculation results of extended mild slope equations. Systematical numerical simulation tests show that the present models can effectively reflect non-linear effects.     

基于强跟踪性能的QEKF组合导航自适应滤波研究

以低成本陆用航姿参考系统（AHRS）/GPS紧耦合系统为研究对象,建立了基于伪距、伪距率、航向角的组合观测数学模型。引入强跟踪滤波（STF）算法,利用渐消因子的作用,增强滤波器对状态快速变化的跟踪能力。同时,针对微惯性器件漂移过大的缺点,采用二阶EKF（QEKF）方法,通过对Hessian矩阵的求解,补偿系统观测方程线性化的二阶截断误差。仿真表明：STQEKF方法可高速准确地逼近系统非线性模型,实现陆地载体导航控制,在传感器精度有限的情况下,使姿态和位置的控制效果较标准EKF分别提高了约8.9%-38.2%和48.7%-54.4%     

利用GPS水准点精化局部CQG2000

CQG2000模型的精度仍不能满足中、大比例尺的测图需要,如何充分利用现有成果,提高其价值和实用性,是目前需要迫切解决的问题。本文提出了通过加密GPS水准点来提高现有CQG2000精度的方法,并研究了GPS水准点数目和分布对提高结果的影响。     

基于分区的局域神经网络时空建模方法研究

区域数据表现为两种尺度的空间特性:反映全局特征的空间依赖性和反映局域特征的空间波动性.空间波动性表现为空间数据在局部地区的聚集或高低交错现象.在研究区域数据时空预测性建模时,从降低数据的空间波动和不平稳性对模型预测能力的影响角度出发,提出了一种基于分区的局域神经网络时空非线性建模的思路.分区过程由基于空间邻接关系的K-means聚类算法完成.不同的分区方案通过相关性、波动性、紧凑性等指标进行评价和优选.在确定最优分区方案的基础上,对各子区分别采用两层前馈网络进行建模,模型的输入不仅要考虑本区内单元的作用,而且要考虑相邻子区的边界效应.各神经网络模型的时空预测能力通过平均相均差和动态相似率等指标进行衡量.最后,通过对法国94个县每周流感报告病例的时空建模分析表明,与全局神经网络模型相比,基于分区的局域神经网络模型具有更好的预测能力.     

GPS数据解算对流层天顶总延迟探讨

文中运用GAMIT/GLOBK 软件,解算了南极长城站GPS接收机天线在不同的相位中心和使用不同星历情况下的天顶总延迟,分析了在不同条件下解算结果的差异,探讨了获取高精度的天顶总延迟和实时获取天顶总延迟的具体方法,为今后实现GPS精密定位和实时天气预报提供了依据.     

捷联惯导MATLAB／Simulink仿真实现

阐述了MATLAB/Simulink的主要特点,设计了基于MATLAB/Simulink的捷联惯性导航仿真模型,给出了仿真实例和分析结果,表明利用MATLAB/Simulink可以提高仿真效率。     

河南理工大学校园导航系统的设计与实现

利用了SuperMap Objects组件,构建了C/S模式的河南理工大学校园导航系统,实现了信息查询、路径导航、办事指南、地图与数据编辑、数据输出、用户管理及三维虚拟校园导航等功能,并提供了与现有校园系统集成的接口.     

北京城市地表温度的遥感时空分析

运用Landsat TM/ETM+和Terra ASTER数据,对北京市1990~2007年夏季的地表温度进行了反演,并对地表温度的空间分布、时间变化作出了分析。对Landsat TM/ETM+数据的温度反演采用了普适性单波段算法,ASTER数据的温度反演采用了劈窗算法。通过对地表温度数据的直方图均衡处理以及综合对比分析,总结出北京地区历年来夏季地表温度的空间分布格局及该格局随北京城市发展的变化规律,分析了研究成果的不足,提出了下一步要努力的方向。