The effects of simulated inundation duration and frequency on litter decomposition: A one-year experiment

There is major uncertainty in the responses of litter decomposition to the inundation regimes in field studies, mainly because of the difficulties in identification of the individual effect of duration and frequency using field studies alone. The interactive role of inundation regime and litter quality also remains unclear. The responses of mass loss to simulated inundation regime (duration and frequency) and litter quality were investigated in leaves of Carex brevicuspis and leaves and stems of Miscanthus sacchariflorus from Dongting Lake, China. Three litter types differing in litter quality were incubated under seven different inundations over 360 days (three single inundations of 90, 180, and 360 days; three repeated 180-day inundations of 2, 3, and 6 times; and no inundations) in a pond near Dongting Lake. Initial N and P contents were highest in C. brevicuspis leaves, intermediate in M. sacchariflorus leaves, and lowest in M. sacchariflorus stems, whereas the organic C, cellulose, and lignin contents were ranked in the opposite order among the three litter types. Decomposition rate was highest in M. sacchariflorus leaves (0.00222–0.00900 day⁻¹), intermediate in C. brevicuspis leaves (0.00135–0.00500 day⁻¹), and lowest in M. sacchariflorus stems (0.00080–0.00100 day⁻¹). The decomposition rate of both C. brevicuspis and M. sacchariflorus leaves increased with increasing inundation duration or decreasing frequency. However, both duration and frequency of inundation had no effect on decomposition of M. sacchariflorus stems. At the end of the incubation, N mineralization was complete in leaf litters with increasing rates with increasing inundation duration or decreasing inundation frequency, but accumulation was found in M. sacchariflorus stems. Organic C decayed quickly in both leaf litters compared with the stem litter. These data indicate that inundation regime has no effect on the decomposition of refractory stem litter while prolonged and stable inundation stimulates the degradation of labile leaf litter.     

2018年斐济MW8.2深震:高频辐射过程和发生原因

2018年8月19日，在斐济东部海域563 km深处发生了一次M_W8.2地震.我们首先挑选位于美国阿拉斯加地区的131个宽频带台站构成台阵，选用垂直分量0.5~2 Hz的高频信号，利用广义台阵反投影技术对这次地震的破裂过程进行了成像，然后基于破裂速度对地震的辐射效率进行了估计.结果表明，这次地震总体上呈单侧破裂，破裂方位在3.0°左右，破裂总长度约51 km，持续时间22 s，平均破裂速度为2.5 km·s^-1.但能量释放有2次高峰，形成两次子事件.第一次为前10 s，峰值在7 s左右，破裂速度为2.9 km·s^-1，辐射效率为45%.第二次为10~22 s，峰值在15 s左右，破裂速度为1.6 km·s^-1，辐射效率为26%.结合震源位置、震源机制、破裂速度以及辐射效率，我们认为这次地震是由于俯冲板块前缘受到下部地幔物质上浮阻力引起的剪切失稳所致，起初板块内部的脆性破裂表现突出，致使辐射效率较高，后来震源处高温高压下的熔融耗散特征逐渐凸现，致使辐射效率下降.     

基于CUDA的地震相干体并行算法

相干体技术在地震勘探资料解释方面得到了广泛的应用,由于相干体技术处理的对象是三维地震数据体,所以算法运算时间较长。为了缩短解释周期,本文充分发挥GPU并行计算优势,对C3相干体算法进行并行化分析。从硬盘读取数据到GPU上计算相干值并写入硬盘的整个过程进行分析,剔除了冗余数据的读取,完成了C3相干体算法的并行化设计与实现。最后分别对串行算法与并行算法进行性能测试,结果表明本文设计的并行算法在保证精度的前提下达到了16倍左右的加速比,对加快地震资料解释具有重要意义。     

大地水准面高的计算及其在海底地形反演中的应用

为充分挖掘海洋重力数据在反演海底地形中的应用潜力,尝试探索利用大地水准面高反演海底地形的技术途径,并以夏威夷—皇帝海山链拐点所在海区作为反演试验区进行验证。首先采用Belikov列推法计算伴随(缔和)勒让德函数,利用EIGEN-6C4地球重力场模型解算获取了分辨率为1'的大地水准面高格网数值模型;然后通过综合分析反演比例函数和转换函数特点、研究海区大地水准面高与海底地形的相干特性以及大地水准面高本身尺度特征,获得了利用大地水准面高反演海底地形的频段范围;最终以试验海区大地水准面高为数据输入,构建了相应的海底地形模型(BNT模型),并与ETOPO1等海深模型进行比对分析。试验结果表明:BNT模型检核差值在一倍均方差范围检核点数量占比70.60%,相比正态分布更加集中;BNT模型检核精度低于ETOPO1等海深模型;海深模型检核精度随着水深增加不断提升,水深小于1 000 m时,海深模型相对误差出现较大发散现象;计算海域ETOPO1模型精度最高,GEBCO模型和DTU10模型检核精度相当。     

海上风电升压站平台高程确定标准的合理性研究

为保证海上风电升压电站建设的经济合理与安全可靠,合理确定海上风电升压电站平台高程十分必要。文中从波浪与潮位的遭遇组合、最大波高取值与现行相关标准的比较、最大波峰高度计算的合理性等方面,全面分析了确定海上风电升压站平台高程各组成项取值标准的合理性,研究认为现行标准明显偏高。建议海上升压站平台底部高程按"100年一遇极端高水位+重现期50年波列累积频率1%的最大波峰高度+安全超高"确定。结合工程实例计算分析,按本文建议可使海上升压站平台高程明显降低,从而节省工程造价,还可减轻升压站工程对周边风机的遮蔽影响,以达到多发电量的效果。     

德令哈13.7m望远镜接收机抗射频干扰研究

在天文观测中,射频干扰会造成假谱,降低数据的可靠性和有效性.射频干扰消减旨在减少干扰信号对射电天文观测的影响,包含器件方面的技术革新和数据处理领域的方法研究.针对德令哈13.7 m望远镜接收机中频部分引入的射频干扰,通过优化中频器件的抗射频干扰能力,提高了接收机的整体抗射频干扰能力,以主动消除方法来减少射频干扰耦合到接收机内部.分析了接收机干扰的传输路径,提出了器件射频干扰的直接耦合系数和器件射频干扰的系统耦合系数的概念,为定位干扰敏感器件并量化干扰引入比重提供了基础.经过抗射频干扰优化后,接收机抗干扰能力改善30 dB左右,望远镜的天文观测效率提高10%以上.     

压裂微震数据初至拾取的时频谱熵法

在油气田开发过程中，微震监测是获得水力压裂引起裂缝分布的一种较为有效的方法。微震的定位成像与裂缝解释需要利用有效微震信号位置，而微震信号具有低信噪比的特点，传统信号拾取方法无法有效实现较低信噪比条件下初至时刻的准确拾取。本文提出一种基于时频谱熵的初至拾取新方法，该方法首先通过S变换获取含噪信号的时频谱；然后对谱内各个采样点沿频率方向进行分帧操作，并计算每帧频段内的近似负熵值，以最小近似负熵值作为该谱点的负熵值；最后沿时间方向比较各谱点的负熵值，最小值对应的时刻即为初至时刻。本文利用不同信噪比的合成地震数据对该方法进行效果验证，并与长短时窗能量比（STA/LTA）法进行拾取结果对比，结果表明：信噪比在-5 dB时，两种方法拾取效果都很好；信噪比在-10 dB时，时频谱熵法拾取效果更好。时频谱熵法更适合低信噪比情况下的信号初至拾取。     

基于小波变换的射频干扰消除方法的研究

在射电天文观测中,射频干扰(Radio Frequency Interference, RFI)会以多种形式混入望远镜接收系统,给观测带来误判或者降低观测信噪比.近年来国内国际射电天文快速发展,国内国际大型射电望远镜和阵列先后建设,观测灵敏度大为提高,射频干扰的影响尤为突出.随着科技发展和人类活动的加剧,射频干扰日益严重且不可逆转.提出利用2维离散小波变换的方法分析射电天文观测的数据,对望远镜系统输出的时间频率序列进行小波变换,根据小波系数分离出原始信号中各分量,每个分量统计得到相应的阈值,将各分量与阈值相比较识别干扰成分并标记去除.利用该方法对实际观测数据进行了处理,结果表明该方法能够很好地标记并消减干扰信号,且提高了观测的信噪比.     

Monitoring and modelling of hydrological processes in the semiarid region of Brazil: The Cariri experimental basins

Catchments have highly variable yields of runoff and soil erosion. The size, land use and the surface cover play a significant role and influence the catchment response and parameter values of simulation models. Two experimental basins—the Cariri basins—were equipped in a semi-arid region of Brazil, for obtaining runoff and sediment yield at different catchment scales, as well as, to evaluate the influence of the land use and surface cover. In the first basin, located in the municipality of Sumé, the field studies were carried out at two different scales: four micro-catchments with an area of around 0.5 ha and nine standard Wischmeier-type erosion plots of 100 m². The experimental units had varied vegetation and management. They were subjected only to natural rainfall events, and were monitored from 1982 to 1991. The total runoff and total sediment yield were determined for each of the events. The installations in the second basin, in the municipality of São João do Cariri, from 1999, include two erosion plots, three micro-catchments, and two sub-catchments of a small basin. These basins are still being monitored for runoff and sediment production. Among the micro-catchments two are nested to detect any scale effect at the micro-catchment level. Nearly 600 events of precipitation, that produced runoff in at least one of the experimental units, have been registered. These data have been used to evaluate the influence of various factors, including cultivation practices and to calibrate hydrological models for plots and micro-catchments. Parameters have been tested by means of cross validations among micro-catchments and sub-catchments. The data sets are made available to all the catchment hydrology researchers and others at https://doi.org/10.5281/zenodo.4690886 .     

Wetland hydroperiod classification in the western prairies using multitemporal synthetic aperture radar

Wetlands represent one of the world's most biodiverse and threatened ecosystem types and were diminished globally by about two‐thirds in the 20th century. There is continuing decline in wetland quantity and function due to infilling and other human activities. In addition, with climate change, warmer temperatures and changes in precipitation and evapotranspiration are reducing wetland surface and groundwater supplies, further altering wetland hydrology and vegetation. There is a need to automate inventory and monitoring of wetlands, and as a study system, we investigated the Shepard Slough wetlands complex, which includes numerous wetlands in urban, suburban, and agricultural zones in the prairie pothole region of southern Alberta, Canada. Here, wetlands are generally confined to depressions in the undulating terrain, challenging wetlands inventory and monitoring. This study applied threshold and frequency analysis routines for high‐resolution, single‐polarization (HH) RADARSAT‐2, synthetic aperture radar mapping. This enabled a growing season surface water extent hyroperiod‐based wetland classification, which can support water and wetland resource monitoring. This 3‐year study demonstrated synthetic aperture radar‐derived multitemporal open‐water masks provided an effective index of wetland permanence class, with overall accuracies of 89% to 95% compared with optical validation data, and RMSE between 0.2 and 0.7 m between model and field validation data. This allowed for characterizing the distribution and dynamics of 4 marsh wetlands hydroperiod classes, temporary, seasonal, semipermanent, and permanent, and mapping of the sequential vegetation bands that included emergent, obligate wetland, facultative wetland, and upland plant communities. Hydroperiod variation and surface water extent were found to be influenced by short‐term rainfall events in both wet and dry years. Seasonal hydroperiods in wetlands were particularly variable if there was a decrease in the temporary or semipermanent hydroperiod classes. In years with extreme rain events, the temporary wetlands especially increased relative to longer lasting wetlands (84% in 2015 with significant rainfall events, compared with 42% otherwise).