浅水多波束测深仪高分辨率波束形成方法研究

常规波束形成技术以其稳健性好、计算量低等特点得到广泛应用,但其空间分辨率受阵元个数限制,不能突破瑞利限。因此,常规波束形成技术应用于高精度浅水多波束测深仪时,存在空间分辨率不足和旁瓣干扰等问题。对比研究了最小方差无失真(MVDR)及多重信号分类(MUSIC)波束形成在浅水多波束测深仪中的应用,给出了浅水多波束测深仪的常规波束形成、MVDR和MUSIC处理方法,并结合能量法和相位法两种底检测测深算法,处理了iBeam8120浅水多波束测深仪外场数据,验证了本文方法的性能。     

顾及周期性误差修正的加权平均温度模型构建

针对目前多数大气加权平均温度(Tm)模型没有考虑季节性影响这一问题,该文首先利用IGRA 2005-2010年全球探空数据,分别建立了各探空站点与地表温度有关的线性Tm模型、与地表水汽压有关的指数Tm模型以及与地表温度和水汽压均有关的混合Tm模型。然后以探空站积分Tm值为参考,对上述3类模型的误差时间序列进行了分析,发现这3种模型均存在周期性误差,并在此基础上构建了考虑周期性误差修正的3类Tm新模型。利用2011-2014年全球探空数据对3类新模型进行精度验证,结果表明:3类Tm新模型的精度相比于原模型均有所提升,模型的周期性误差影响基本得以消除,且3类Tm新模型的精度基本一致。     

基于改进BP神经网络的海底底质分类

通过采用遗传算法优化神经网络初始权值的方法,将GA算法与BP神经网络有机结合,应用于海底底质分类。基于多波束测深系统获取的反向散射强度数据,应用改进的BP神经网络分类方法,实现对海底基岩、砾石、砂、细砂和泥等底质类型的快速、准确识别。通过实验比较,GA-BP神经网络分类精度明显高于BP神经网络,证明了该方法的有效性和可靠性。     

多波束测深中声速剖面的横向加密方法

针对声速剖面在横向空间上的低密度、分布不均匀导致水深测量误差的问题,提出了一种容易实现且有效的思路,采用分层取样、空间插值的方法,在后处理过程中人为地增加声速剖面的数量,使其密度足够大、分布更均匀,达到改善声速改正的效果和提高多波束测深精度的目的。     

遥感水深反演海图修测应用研究

卫星遥感技术可以应用于浅海水下目标探测,是特殊区域碍航物探测的有效手段。结合海岛礁的特殊地理环境和测绘需求,采用高分辨率卫星影像进行浅水深反演试验,提出的高分辨率预处理方法实用、有效;并与海图数据结合,进行水深标定和海图修测应用分析。研究结果表明,针对南海岛礁,应用高分辨率卫星影像可以获取岛礁分布和轮廓、概略等深线信息、潜在航道以及水下障碍物信息等,能达到修测海图的目的。     

基于声波探测海底浅层沉积物特性的方法研究

利用声波在水中的传输特性进行海底浅层沉积物特性的探测。在实践应用中有实验室方法、现场测量方法和海底声学遥测方法,并研制出大量的仪器。由于海上和海底实际条件的限制,声波探测方法都存在一定的不足之处,由此提出了声衰减测试沉积物特性以及建立实验室模拟海底环境探测方法的设想。     

东海陆架两期沙脊的时空对比东海陆架两期沙脊的时空对比

基于高分辨率的单道地震和多波束测深数据,识别并对比了东海陆架中部同一海区相距20余万年的层U14和层U2两期沙脊群,其中层U14期沙脊属于埋藏沙脊,位于东海海底以下90 m深处,推测属于距今320~200 ka的海侵体系域(TST),沙脊顶界面是该期海侵的最大洪泛面(MFS);层U2期沙脊位于东海陆架,属于衰退沙脊,系末次盛冰期(LGM)以来的TST,顶界面是LGM以来的MFS。尽管两期沙脊形成年代相距20余万年,地层层位相距近90 m,但是沙脊群总体走向一致,表明距今2×105 a以来东海陆架潮波基本格局稳定。从层U2期可识别出4个亚期沙脊,通过多波束海底地形图可识别出4组走向的沙脊,多亚期、多走向沙脊是LGM以来海平面阶梯状波动在海底地形演变过程中的响应证据。     

主检测深线系统误差的探测与实现

针对海道测量中测深成果系统误差的评价问题,介绍了一种利用主、检测深线测深数据探测系统误差的方法,以实现测深中系统误差的科学评估。     

DGPS测深技术在海底管道工程中的应用

简要介绍了差分全球定位系统(DGPS)的基本定位原理、DGPS测深技术中水深测量系统的硬件组成。重点分析了福建LNG站线湄洲湾海底管道工程成功采用DGPS测深技术进行勘测作业,使测量精度和工程效率得到极大地提高。这一工程实践经验说明,DGPS测深技术逐步为海底管道工程所采用,但其是一个不断探索、不断完善的过程,对今后我国沿海地区的海底管道工程的勘测建设具有一定的借鉴作用。     

Channelled erosion through a marine dump site of dredge spoils at the mouth of the Puyallup River, Washington State, USA

Channels are relatively common on river-mouth deltas, but the process by which they arise from river sediment discharge is unclear because they can potentially be explained either by negatively buoyant (hyperpycnal) flows produced directly from the river outflow or by flows generated by repeated failure and mobilisation of sediment rapidly deposited at the delta front. Channels eroded through a dump site of dredge spoils are described here from multibeam and older sonar data collected in Commencement Bay, at the mouth of the Puyallup River. Shallow channels on the seaward upper surface of the dump site, away from any flows that could have been produced by delta front failures, suggest that at least some hyperpycnal flows were produced directly from the positively buoyant river outflow up to 200 m from the edge of the river mouth platform. The form of channel bed erosion is revealed by the longitudinal shape of the main eroded channel compared with the adjacent dump site profile. It suggests that the channel evolved by its steep front retreating, rather than by simple vertical entrenchment or diffusive-like evolution of the profile, a geometry interpreted as evidence that repeated failure of the bed occurred in response to shear stress imposed by bottom-travelling flows. Model calculations based on shear strengths back-calculated from the geometry of channel wall failures suggest that, if the main channel were eroded solely by hyperpycnal flows, their generation was remarkably efficient in order to create flows vigorous enough to cause channel bed failure. Besides the sediment concentration and discharge characteristics that have been considered to dictate the ability of rivers to produce hyperpycnal flows, it is suggested that the timing of floods with respect to the tidal cycle should also be important because extreme low tides may be needed to ensure that coarse sediment is transferred vigorously to the edge of river mouth platforms.