稻田土壤中甲烷产生率的实验研究

本实验旨在研究稻田土壤中甲烷产生率对稻田CH_4排放的影响.观测结果表明:土壤各深度的甲烷产生率有很大的变化范围(1—4639ng·h~(-1)·g~(-1)d.w.).主要的甲烷产生区域是7—17cm深的土壤层,其中以13cm深的土壤层上的生成速率最大.土壤中甲烷产生率与稻田CH_4排放率在水稻生长的某些阶段有较好的相关性,但它的季节变化却不能与排放的季节变化完全耦合.在水稻生长期,土壤中甲烷产生率随时间而增大,并在8月份水稻收割前达到最大,其日平均值在38—767 ng·h~(-1)·g~(-1)d.w.间变动.稻田土壤中甲烷产生率也存在日变化,一般在下午达到最大值,但却没有发现它与土壤温度有明显的相关关系.在不同施肥及水稻品种的稻田土壤中也观测到不同的甲烷产生率.在土壤中产生的甲烷最多只有28.8％被排放到大气中,而其余多于71.2％的则被氧化在土壤中.     

Study of interplanetary magnetic field with Ground State Alignment

We demonstrate a new way of studying interplanetary magnetic field—Ground State Alignment (GSA). Instead of sending thousands of space probes, GSA allows magnetic mapping with any ground telescope facilities equipped with spectropolarimeter. The polarization of spectral lines that are pumped by the anisotropic radiation from the Sun is influenced by the magnetic realignment, which happens for magnetic field (<1 G). As a result, the linear polarization becomes an excellent tracer of the embedded magnetic field. The method is illustrated by our synthetic observations of the Jupiter’s Io and comet Halley. Polarization at each point was constructed according to the local magnetic field detected by spacecrafts. Both spatial and temporal variations of turbulent magnetic field can be traced with this technique as well. The influence of magnetic field on the polarization of scattered light is discussed in detail. For remote regions like the IBEX ribbons discovered at the boundary of interstellar medium, GSA provides a unique diagnostics of magnetic field.     

POS辅助下的车载序列影像匹配

针对车载序列影像中所包含的具有大量重复纹理的建筑立面区域给现有匹配方法带来的误匹配率高、匹配点不足等不适应性问题,提出了一种联合POS辅助及KLT（Kanade-Lucas-Tomasi）角点特征跟踪的新匹配方法。通过充分利用POS信息所能够建立的局部近似建筑立面分层平面及DEM信息,获取KLT跟踪算法中所需的初始位移信息,并将建筑立面及路面区域进行分开追踪匹配,以保证匹配点的良好分布特性,最后利用RANSAC结合核线约束进行误差剔除得到最终匹配结果。试验证明本文方法能够有效解决车载序列影像中重复纹理区域的匹配问题,可高效获取分布均匀、数量充足的匹配特征点,为后期空中三角测量提供更稳健的同名约束条件。     

Identification of ice elevation change of the Shuiguan River No. 4 glacier in the Qilian Mountains,China

GPS measurement,an effective method for surveying glacier surface topography,has been applied in some glaciers for many years.The Shuiguan River No.4 glacier,a small glacier with its area of 1.84 km2 in 1972,located in the east of the Qilian Mountains,China,was selected to study its ice elevation change using GPS measurement in 2007.This study was conducted on the ablation area with GPS-measured area 0.5 km2.The ice elevation change of the glacier was obtained by comparing the DEM obtained by a 1:50 000 topographic map made in 1972 with the DEM by GPS-measured data acquired in 2007.The differences of the two DEMs showed the thinning condition of the glacier was apparent.The mean thinning was 15±8 m with the mean thinning rate of 0.42±0.22 m a-1 for 1972-2007 in the measurement area,which equaled 0.38±0.20 m yr-1 in water equivalent(w.e.).The prominent thinning occurred on the south part of the glacier,which was the area near the glacier terminus with the maximum thinning of 41±8 m.Assuming the thinning value of 15±8 m for the glacier area below 4640 m a.s.l.,the wasting ice mass was calculated to be 6.4±3.2×10-3 km3 for 1972-2007,corresponding to 5.7±2.8×10-3 km3 w.e.,which meant that the montane runoff released by the glacier was at least 5.7±2.8×106 m3 between 1972-2007.     

山西省湿地生物多样性及其保护

山西省共有各类湿地２３０９ｋｍ＾２,包括河口（内陆）湿地、河流湿地、湖泊湿地、水库湿地、沼泽和草甸湿地等,主要分布于各河流流域、湖泊和水库及周围地区。山西湿地有着丰富的生物多样性,其中动物４５５种、植物１２０９种,植被类型有７０个群系。此外,还分析了山西湿地生物多样性受威胁的现状和原因,提出了山西湿地生物多样性保护对策。     

稻田CH_4排放的控制措施

通过对稻田CH_4排放施肥效应的研究指出:化肥的施用能够降低CH_4的排放,但是有机物的数量及质量是影响稻田CH_4排放的主要因素,因此稻田CH_4排放的控制应该主要从有机肥的科学施用入手;比起常规的有机肥,沼渣肥这种已经发酵的“陈”有机肥能够较大程度地降低CH_4排放。推出沼渣肥和化肥混施的方案,在不降低水稻产量的同时降低CH_4排放,并建议新鲜有机肥(紫云英、稻草、猪粪等)先进沼气池产生沼气并利用,沼渣在施入土壤前先充分干燥。 水管理实验发现,当土壤湿润度低到一定程度时(26％<湿度<31％),CH_4排放率突然减少,CH_4产生率也明显降低,而且主要的产CH_4区域向土壤深处移动。重新灌水后在较长的一段时间内CH_4排放率仍不能恢复。一个三日间隔灌溉法因为灌溉的时间间隔太短,没有起到降低排放的作用。同时考虑水稻产量及方案实用性,提出用适当时间表的间歇灌溉来降低CH_4排放。如果能控制好土壤湿润度的临界值,我国常用的晒田技术会起到降低CH_4排放的效果。 为更好地降低CH_4排放,并努力增加水稻产量,设想了一种水肥结合的控制措施,即把沼渣肥和化肥混施方案与最简单的间歇灌溉方案——晒田共同使用、结果使晚稻CH_4排放降低了一倍多。水稻作为传输CH_4的主要路径,某些水稻品种也可能对降低稻田CH_4排     

基于二元LSTM神经网络的船动积分位移预测算法研究

摘要：在海况环境下,进行船舶运动预测时。由于惯性传感器采集系统本身的电学特性,会产生偏移误差,严重影响一般预测方法的准确性。针对这一问题,在常规LSTM神经网络的基础上,设计改良了一种二元的LSTM网络架构。在船舶运动仿真平台上进行模拟船舶升沉运动实验,并通过惯性传感系统测量仿真平台实时积分位移进行计算验证。验证统计该网络预测结果峰差值均方差0.64%,均值均方差0.42%,峰值均方差0.57%,证实该网络较常规LSTM在船舶运动预测领域具有更好的针对性和适应性,更准确的还原预测实际的船舶运动轨迹。     

Coupling a glacier melt model to the Variable Infiltration Capacity (VIC) model for hydrological modeling in north-western China

For the sustainable utilization of rivers in the mid and downstream regions, it is essential that land surface hydrological processes are quantified in high cold mountains regions, as it is in these regions where most of the larger rivers in China acquire their headstreams. Glaciers are one of the most important water resources of north-west China. However, they are seldom explicitly considered within hydrological models, and climate-change effects on glaciers, permafrost and snow cover will have increasingly important consequences for runoff. In this study, an energy-balance ice-melt model was integrated within the Variable Infiltration Capacity (VIC) macroscale hydrological model. The extended VIC model was applied to simulate the hydrological processes in the Aksu River basin, a large mountainous and glaciered catchment in north-west China. The runoff components and their response to climate change were analyzed based on the simulated and observed data. The model showed an acceptable performance, and achieved an efficiency coefficient R ² ≈ 0.8 for the complete simulation period. The results indicate that a large proportion of the catchment runoff is derived from ice meltwater and snowmelt water. In addition, over the last 38 years, rising temperature caused an extension in the snow/ice melting period and a reduction in the seasonality signal of runoff. Due to teh increased precipitation runoff, the Aksu catchment annual runoff had a positive trend, increasing by about 40.00 × 10⁶ m³ per year, or 25.7 %.     

多环芳烃油指纹应用于船舶溢油鉴别研究

溢油种类主要包括船舶燃料油和原油,二者性质的差异决定了鉴别方法也相应不同,寻求适合于船舶溢油的鉴别方法具有重要意义。在使用柱色谱层析方法对样品进行分离前处理的基础上,以气相色谱/质谱方法(GC-MS)为主要分析手段,对溢油样品和可疑船舶溢油源样品的多环芳烃油指纹特征进行对比,并在多环芳烃油指纹参数的基础上进一步进行多环芳烃内组成三角图分布特征与聚类分析研究,成功为珠江口水域某船舶溢油事故追踪到肇事溢油源。结果表明:取自丁船的油样和现场溢油样芳烃油指纹特征最为相近,是此次溢油事故的溢油源。受风化作用后的船舶燃料油中饱和烃类化合物数量稀少,且含量极低,不适合用于溢油鉴别,而多环芳烃类化合物较饱和烃类化合物而言具有含量高、种类丰富的特点,是该类溢油鉴别的主要油指纹依据。使用油指纹参数进行可疑溢油源识别时,充分考虑油品中有机分子所受风化影响程度的不同是风化条件下溢油鉴定的关键。因此,多环芳烃油指纹可以有效应用于船舶燃料油溢油的鉴别。