地震观测传递函数的精度问题

数字地震观测的传递函数由反馈地震计和地震数据采集器的传递函数合成。地震数据采集器的传递函数主要用数字算法生成,其精度可以做得很高。反馈地震计是机电一体化的模拟传感部件,其传递函数的精度和稳定性均欠佳。尤其是由反馈地震计生成的主导二阶极点,对应传统地震观测的周期和阻尼两个重要参数,其精度和稳定性往往较低,不利于地震学研究深入发展的需要。以四川地震观测的标定数据为例,说明数字地震观测传递函数的精度和稳定性急待改进。     

基于驱动脉冲延迟的线阵CCD亚像素测量系统

通过CPLD设计间隔精准延迟的驱动脉冲,将线阵CCD空间信息转换为在一个脉冲单元内变化的时间信息,并通过高精度步进电机位移平台进行了检测,实验证明该方法可以有效地突破线阵CCD像元精度限制,实现亚像素测量。     

Assessing the role of forests in mitigating eutrophication downstream of pasture during spring snowmelt

Little research has examined whether forests reduce stream water eutrophication in agricultural areas during spring snowmelt periods. This study evaluated the role of forests in ameliorating deteriorated stream water quality in agricultural areas, including pasture, during snowmelt periods. Temporal variation in stream water quality at a mixed land‐use basin (565 ha: pasture 13%, forestry 87%), northern Japan, was monitored for 7 years. Synoptic stream water sampling was also conducted at 16 sites across a wide range of forest and agricultural areas in a basin (18.3 km²) in spring, summer and fall. Atmospheric nitrogen (N) and phosphorus (P) deposition were measured for 4 years. The results showed that concentration pulses of nitrate, organic N and total P in stream water were observed when discharge increased during spring snowmelt. Their concentrations were high when silicate concentrations were low, suggesting surface water exported from pasture largely contributed to stream water pollution during snowmelt. Atmospheric N and P deposition (4.1 kg N ha^?1 y^?1; 0.09 kg P ha^?1 y^?1, respectively) was too low to affect the background concentrations of N and P in streams from forested areas. Reduction of eutrophication caused by nutrients from pasture was mainly due to dilution by water containing low concentrations of N and P exported from forested areas, whereas in‐stream reduction was not a dominant process. Results indicate that forests have a limited capacity to reduce the concentration pulses of N and P in stream water during snowmelt in this study basin. Copyright © 2014 John Wiley & Sons, Ltd.     

桥梁变形监测数据小波去噪与Kalman滤波研究

针对跨海大桥三维变形监测问题,实施了测量机器人与三维激光扫描仪测量方案,二者三维监测数据最大较差不超过0.5 mm,而且测量机器人监测数据精度高,说明三维激光扫描技术应用于桥梁变形监测是可行的。根据小波分析与Kalman滤波理论,建立小波去噪后的桥梁变形监测Kalman滤波模型。实验表明,经过小波去噪的监测数据再进行Kalman滤波处理,提高了桥梁变形预测与变形分析的可靠性。通过标准Kalman滤波、自适应Kalman滤波桥梁变形观测数据处理,及平方和误差、均方误差、平均相对误差精度分析知,自适应Kalman滤波要优于标准Kalman滤波方法。     

三度体重力矢量的有限单元法正演计算

以重力位在场源内部满足泊松方程为依据,以重力矢量满足第三类边界条件为切入点,推导了与三度体重力矢量满足的边值问题相对应的变分问题,进而利用有限单元法实现了对变分问题的求解.立方体模型试验结果表明:文中提出的新的系数矩阵存储方式较之传统方式能够更有效地节约存储空间,且为利用预条件共轭梯度技术更加快速地求解线性方程组提供了保障;重力矢量的计算精度与边界长度及单元网格的边长息息相关,其计算效率则主要取决于所要计算的节点总数和大型稀疏线性方程组求解算法的优劣;一般情况下,当单元的边长小于场源体边长的1/10、边界长度大于场源体长度的7.5倍时,能够获得理想的结果.     

Milking the Plains: movement of large dairy operations into southwestern Kansas

Western Kansas has an historical identification with cattle, with a focus on cattle ranching and more specifically since the 1950s, beef-cattle feedlots. Since the mid-1990s large dairy operations have moved into southwestern Kansas. Today more than twenty large dairies house more than 70,000 milk cows. These operate as confined feeding operations similar to beef-cattle feedlots. Regional advantages for the dairy industry include affordable land with wide-open space, local residents' cattle- and dairy-friendly attitudes, and other factors. Regional promoters have actively recruited dairies, and a dairy-business support system has emerged. The prospects for continued expansion of dairies in southwestern Kansas are unclear; despite the locational advantages and the possibility that the industry may continue to relocate here, as did the cattle-feeding industry several decades ago, further moves into the area may depend on continued resources availability and additional infrastructure development.     

MEXICAN URBAN COLONIAS IN THE SALT RIVER VALLEY OF ARIZONA

The history of the Salt River Valley is often characterized as a boomtown narrative that highlights the exploding Anglo population and rapid suburban sprawl that followed World War II. Frequently overlooked is the fundamental role that Mexicans played in the early suburbanization process through the formation of urban colonias, or suburban Mexican settlements. Early Mexican suburbs, despite their rural atmosphere, were connected to the pre–World War II urbanization process in the Salt River Valley. Our research seeks to amend and clarify the urban historical record by documenting the diverse and dispersed array of Mexican suburban settlements, a process generated by industrial agriculture, decentralization, and residential segregation.     

江苏省典型样带区农业发展的地域差异

以江苏省204国道沿线15县(市)组成的典型样带为研究区域,根据农林牧渔业总产值及其增长率、农业产值占地区GDP比重、农业产业结构、农业优势度等多项发展指标,应用统计模型和GIS空间分析方法,研究了改革开放以来样带区农业发展的差异格局。研究结果表明:(1)近30年来样带内各县市农业发展的总体水平差距拉大,农林牧渔业总产值年均增长率呈现出北高南低的格局;(2)农业产值占地区GDP比重普遍下降,尤以苏南三市下降幅度最大,样带内各县市间农业比重的差距明显增大;(3)样带内各县市农业内部结构也发生了巨大的变化,其中种植业比重有所下降,渔业比重有所提高,并且各县市间种植业比重的差距和渔业比重的差距均有明显增大;(4)改革开放以来,样带区农业优势度北高南低的格局得到逐渐强化,苏南三市不具优势,苏北地区优势度最高。     

样本数据预处理对基于BP神经网络的GPS高程拟合的影响

分析在基于BP神经网络的GPS高程拟合建模中样本数据预处理的必要性,并列举了归一化、中心化、标准化3种数据预处理方法,然后结合实例,在神经网络建模中增加一个数据预处理层,分别用3种训练方法对基于不同数据预处理的模型进行训练建模,将计算结果进行对比分析,并与二次曲面模型结果进行比较,得出不同数据预处理方法对基于神经网络的GPS高程拟合建模精度的影响不同,且神经网络方法比二次曲面方法的拟合精度更高。     

基于非差与历元差分两种观测模型估计精密卫星钟差的方法比较

研究了非差与历元差分两种观测模型估计精密卫星钟差的方法,评价了分别利用两类观测模型估钟的特点。通过实际算例分析了两种观测模型估钟的处理速度与精度。计算结果表明：基于非差观测模型估计卫星钟差精度高、观测信息没有损失、可靠性高、可以实现模糊度固定,但由于未知参数多,解算速度较慢,且需要经过一段时间的收敛才能达到所需精度;而历元差分模型估计卫星钟差待求参数较少,计算效率高,且不存在收敛过程,但估钟精度比非差模型估钟收敛后的精度略低,且得不到钟差初值,需从导航电文中提取或通过其他方式获取,不过由此引起的系统性偏差,在定位时可被模糊度和接收机钟差吸收,不影响最终的定位结果。