-
研制的偏振海洋激光雷达系统主要由发射系统、接收系统以及数据采集和控制系统三大部分组成,系统结构示意图如图1所示[21]。发射系统主要为一台倍频调Q的Nd:YAG激光器,产生的激光波长为532 nm,脉冲能量为5 mJ,脉宽约为10 ns,重复频率为10 Hz。激光由激光器出射并经准直扩束后,通过起偏器获得高消光比的线偏振光。接收系统包含平行和正交偏振两个通道,由望远镜、线偏振片、光阑、干涉滤光片、准直透镜、光电探测器等组成,分别用于收集平行和垂直于出射激光偏振方向的后向散射光。正交与平行偏振通道信号的强度比值被称为退偏比,退偏比与散射物质的表面形状、粒径大小、折射率等密切相关,可用于散射物质的识别与分类[16]。此外,还设置了视频监控通道用于实时监控激光入射点周围的海况,如海浪大小、环境光变化、海面垃圾、近表层海洋生物等。数据采集和控制系统由高速采集卡和计算机组成。
研制的偏振海洋激光雷达系统,于2017年8月搭载于“海力”号科考船在黄海海域开展了为期20多天的海试实验,图2为海试航线图。所研究的海域主要包括近岸的平山岛水域(缩写为PSD,图2中红色),远岸的冷水团水域(缩写为LST,图2中黄色),以及靠近渤海的威海水域(缩写为WH,图2中蓝色)。
-
数据预处理主要包括降噪和对齐海面。降噪指对激光雷达信号廓线进行平均和去除背景噪声。由于检测目标是在水中运动的水母,属于强散射个体,区别于反演水体光学特性或检测浮游植物层次常选择100条甚至更多廓线进行平均以得到较为平滑的廓线[22],这里平均次数降到了10,即每1 s统计一条廓线。计算背景噪声的方法为对每一条廓线最大探测深度以下的采样点信号强度进行平均。
对齐海面的关键是找到各条廓线的海面位置。正交偏振通道信号更容易受到海面附近水雾的影响而导致检测到错误的海面,利用平行偏振通道信号检测海面位置更为准确。对于每一条廓线,海面峰前都存在一个大气峰,这两个峰的位置在不同廓线中都比较稳定,仅在小范围内波动,为了避免大气峰的干扰,检测海面位置的方法是在一个固定的范围内找到信号最强点。在一些特殊情况下,譬如海面以上存在海鸟、飞虫、水花、泡沫等,会出现海面错检。因此在完成海面检测后还需对突变较大的海面进行校正,最后将所有廓线的海面位置对齐。
-
数据预处理后,对正交偏振通道的信号廓线从海面至最大探测深度进行峰凸起检测。在对数坐标系下,理论上信号是线性下降的,当其中存在一个明显的峰凸起时,说明此处存在强散射信号,可能来源于强散射个体或群体。假如峰凸起持续廓线数量很少,如图3(a)中峰凸起廓线前后5条廓线都是理想的线性下降,说明该峰凸起来源于强散射个体,对应图3(b)存在零散个体信号。倘若峰凸起持续廓线数量多且深度几乎相同,如图3(c)中峰凸起前后20条廓线(甚至可能成百上千条廓线)在同一深度处都存在峰凸起,则该峰凸起来源于强散射群体,对应图3(d)中的强散射层次。检测峰凸起的方法是差分遍历廓线中的所有数据,依次找到所有峰凸起,并保存峰值及深度、上升沿及下降沿等特征信息。
图 3 强散射个体和强散射群体的信号区别。 (a)峰凸起持续廓线少;(b)存在强散射个体;(c)峰凸起持续廓线多;(d)存在强散射层次
Figure 3. Signal difference between strongly scattered individual and strongly scattered population. (a) Few continuous profiles with peak bulge; (b) Exist strongly scattered individual; (c) Many continuous profiles with peak bulge; (d) Exist strongly scattered layer
-
由于数据预处理中只进行了10次廓线平均,平均后的廓线仍不够平滑,峰凸起检测结果既包含强散射个体,即明显且持续时间短的峰凸起,又存在许多干扰峰凸起,譬如噪声引起的信号波动峰、强散射层次(如浮游植物层)引起的持续时间长的峰凸起等,因此需要对所有峰凸起进行筛选工作。筛选依据为以下三个指标:峰凸起形状、峰凸起强度、峰凸起对比度。
峰凸起形状筛选通过设置上升沿和下降沿包含的最少数据点,可以剔除绝大部分噪声引起的锯齿峰。峰凸起强度的筛选规则为设置峰凸起强度阈值,从而剔除噪声导致的信号波动峰。峰凸起强度的定义为峰值强度与上升沿起始点强度之差。由于信号波动较为剧烈,上升沿起始点可能位于上一个峰凸起的谷值而导致低估,解决方法是用周围连续一段廓线在该上升沿起始点的强度平均值替代。此外,由于廓线信号强度随深度指数衰减,峰凸起强度的阈值不能简单设置为一个常数,而应达到自适应的效果,解决方法是基于周围连续一段廓线在此处的峰凸起强度设置合适的阈值。
峰凸起对比度定义为峰凸起信号分量与背景信号分量的比值。背景信号分量近似用周围连续一段廓线的信号平均值表示。类似峰凸起强度,设置峰凸起对比度的自适应阈值。若峰凸起来源于强散射个体,则背景信号为水体,得到的峰凸起对比度大于1;若凸起来源于以强散射层次为代表的强散射群体,则背景信号仍为该强散射群体,峰凸起对比度约为1。因此,通过峰凸起对比度条件筛选能够有效避免强散射层次的干扰,只保留来自强散射个体的峰凸起,如图4所示。
-
为了定量描述检测到的强散射个体的光学特性,以水母为例,定义了退偏率和信号对比度两个光学参数。
水母退偏率depoj的计算见公式(1),其中Sx、Sp分别为数据预处理后的正交和平行偏振通道信号在峰凸起峰值位置的信号强度,下标j,w分别表示信号中的水母分量和背景水体分量,如图5所示。关键问题是分离水母信号和背景水体信号,解决方法是在对数坐标下选取合适的范围进行线性拟合。
图 5 线性拟合法分离背景水体信号和水母信号
Figure 5. Separate the jellyfish signal and the background water signal by linear fitting method
$$dep{o_{\rm{j}}} = \frac{{{S_{{\rm{x}},{\rm{j}}}}}}{{{S_{{\rm{p}},{\rm{j}}}}}} = \frac{{{S_{\rm{x}}} - {S_{{\rm{x}},{\rm{w}}}}}}{{{S_{\rm{p}}} - {S_{{\rm{p}},{\rm{w}}}}}}$$ (1) 信号对比度Contrast的定义为后向散射信号中的水母分量与背景水体分量的比值,见公式(2),它反映了在激光雷达后向散射回波信号中水母与海水的对比度,Contrast数值越大,说明越容易观察到海水中的水母。
$$Contrast = \frac{{{S_{{\rm{x}},{\rm{j}}}} + {S_{{\rm{p}},{\rm{j}}}}}}{{{S_{{\rm{x}},{\rm{w}}}} + {S_{{\rm{p}},{\rm{w}}}}}}$$ (2)
Characteristics of jellyfish in the Yellow Sea detected by polarized oceanic lidar
-
摘要:
近年的研究表明:全球的许多海湾和海区出现水母数量增加甚至暴发的现象,对海洋生态环境、海洋渔业、滨海旅游业、核电安全等产生了负面影响。2017年8月,自主研制的船载偏振海洋激光雷达在黄海海域开展了实验测量,观测到丰富的强散射个体信号,结合视频监控信息,判断信号来源于水母(沙海蛰),证明偏振海洋激光雷达能够实现水母个体的遥感探测。研究结果表明,同一水域水母的光学特性表现出聚类的特点;不同水域的水母信号对比度相近、退偏率不同,说明水母的光学特性与其生存水域环境密切相关。因此,偏振海洋激光雷达可以高效、经济、精确监测水母分布和数量变动状况,其未来的推广可以完善我国海域水母动态监测手段。
Abstract:Recent studies have shown an increase or even outbreaks of jellyfish in many bays and seas around the world, which has a negative impact on the marine ecological environment, marine fisheries, coastal tourism, nuclear safety and so on. In August 2017, the self-developed shipborne polarized oceanic lidar carried out experimental measurements in the Yellow Sea. Rich strongly scattered individual signals were observed. Combined with the video monitor information, the signals were determined to come from the jellyfish (Nemopilema nomurai), demonstrating the polarized oceanic lidar is available to realize the remote sensing detection of jellyfish. The results show that the optical properties of jellyfish in the same waters show clustering. The signal contrast distribution of jellyfish in different waters was similar, and the distribution of depolarization rate was different, indicating that the optical properties of jellyfish were closely related to the water environment where they lived. As a result, the polarized oceanic lidar can monitor the distribution and population changes of jellyfish efficiently, economically and accurately and its future promotion can improve the dynamic monitoring methods of jellyfish in Chinese waters.
-
图 3 强散射个体和强散射群体的信号区别。 (a)峰凸起持续廓线少;(b)存在强散射个体;(c)峰凸起持续廓线多;(d)存在强散射层次
Figure 3. Signal difference between strongly scattered individual and strongly scattered population. (a) Few continuous profiles with peak bulge; (b) Exist strongly scattered individual; (c) Many continuous profiles with peak bulge; (d) Exist strongly scattered layer
-
[1] Doyle T K, Houghton J, Buckley S M, et al. The broad-scale distribution of five jellyfish species across a temperate coastal environment [J]. Hydrobiologia, 2007, 579(1): 29-39. doi: 10.1007/s10750-006-0362-2 [2] Brodeur R D. Increases in jellyfish biomass in the Bering Sea: implications for the ecosystem [J]. Marine Ecology Progress, 2002, 233(4): 89-103. [3] Uye S I. Blooms of the giant jellyfish Nemopilema nomurai: A threat to the fisheries sustainability of the East Asian Marginal Seas [J]. Plankton Benthos Research, 2008, 3(S): 125-131. doi: 10.3800/pbr.3.125 [4] Dong Z, Liu D, Keesing J K. Jellyfish blooms in China: Dominant species, causes and consequences [J]. Marine Pollution Bulletin, 2010, 60(7): 954-963. doi: 10.1016/j.marpolbul.2010.04.022 [5] Zhang F, Li C L, Sun S, et al. Progress on studying jellyfish bloom, and the monitoring and control [J]. Oceanologia et Limnologia Sinica, 2017, 48(6): 1187-1195. (in Chinese) [6] Purcell J E. Predation on zooplankton by large jellyfish, Aurelia labiata, Cyanea capillata and Aequorea aequorea, in Prince William Sound, Alaska [J]. Mar Ecol Prog Ser, 2003, 246: 137-52. doi: 10.3354/meps246137 [7] Broduer R D, Mills C E, Overland J E, et al. Evidence for a substantial increase in gelatinous zooplankton in the Bering Sea, with possible links to climate change [J]. Fisheries Oceanography, 2010, 8(4): 296-306. [8] Zuo T, Wang J, Wu Q, et al. Spatical distribution and biomass of large jellyfish in the Yellow Sea and northern part of the East China Sea in May 2015 [J]. Oceanologia et Limnologia Sinica, 2016, 47(1): 195-204. (in Chinese) [9] Graham W M, Martin D L, Martin J C. In situ quantification and analysis of large jellyfish using a novel video profiler [J]. Marine Ecology Progress, 2003, 254: 129-140. [10] Houghton J, Doyle T K, Davenport J, et al. Developing a simple, rapid method for identifying and monitoring jellyfish aggregations from the air [J]. Marine Ecology Progress Series, 2006, 314(1): 159-170. [11] Mano T, Guo X, Fujll N, et al. Moon jellyfish aggregations observed by a scientific echo sounder and an underwater video camera and their relation to internal waves [J]. Journal of Oceanography, 2019, 75(4): 359-374. doi: 10.1007/s10872-019-00507-8 [12] Zhang J, Zhang S, An C, et al. An effective detection method based on the biological acoustic characteristics of the outlet of nuclear power plant [J]. IOP Conference Series Materials Science, 2020, 780: 022034. doi: 10.1088/1757-899X/780/2/022034 [13] Wang B, Fang L C, Dong J, et al. Review of acoustic techniques in the monitoring and assessment of the giant jellyfish [J]. Acta Ecologica Sinica, 2017, 37(24): 8187-8196. (in Chinese) [14] Shin H H, Han I, Oh W, et al. Estimation of moon jellyfish Aurelia coerulea using hydroacoustic methods off the coast of Tongyeong, Korea [J]. Korean Journal of Fisheries Aquatic Sciences, 2019, 52(6): 725-734. [15] Hewitt R P, Watkins J, Naganobu M, et al. Biomass of Antarctic krill in the Scotia Sea in January/February 2000 and its use in revising an estimate of precautionary yield [J]. Deep Sea Research Part II: Topical Studies in Oceanography, 2004, 51(12-13): 1215-1236. doi: 10.1016/S0967-0645(04)00076-1 [16] Vasilkov A P, Goldin Y A, Gureev B A, et al. Airborne polarized lidar detection of scattering layers in the ocean [J]. Appl Opt, 2001, 40(24): 4353-4364. doi: 10.1364/AO.40.004353 [17] Churnside J H. Polarization effects on oceanographic lidar [J]. Optics Express, 2008, 16(2): 1196-1207. doi: 10.1364/OE.16.001196 [18] Churnside J H. Review of profiling oceanographic lidar [J]. Opt Eng, 2014, 53(5): 051405. [19] Churnside J H, Wilson J J, Tatarskii V V. Lidar profiles of fish schools [J]. Appl Opt, 1997, 36(24): 6011-6020. doi: 10.1364/AO.36.006011 [20] Churnside J H, Marchbanks R D, Donaghay P L, et al. Hollow aggregations of moon jellyfish (Aurelia spp.) [J]. Journal of Plankton Research, 2016, 38(1): 122-130. doi: 10.1093/plankt/fbv092 [21] Zhou Y D, Liu D, Xu P T, et al. Detecting atmospheric-water optical property profiles with a polarized lidar [J]. Journal of Remote Sensing, 2019, 23(1): 108-115. (in Chinese) [22] Xu Peituo, Tao Yuting, Liu Zhipeng, et al. Comparison of oceanic lidar experiments and simulation results [J]. Infrared and Laser Engineering, 2020, 49(2): 0203007. (in Chinese) [23] Ding F Y, Cheng J H. Dynamic distribution of Stomolophus meleagris in the East China Sea Region [J]. Journal of Fishery Sciences of China, 2007, 14(1): 83-89. (in Chinese) [24] Zhang F, Sun S, Li C L. Estimation on food requirement by large jellyfish Nemopilema nomurai in summer [J]. Oceanologia et Limnologia Sinica, 2017, 48(6): 1355-1361. (in Chinese)
计量
- 文章访问数: 532
- HTML全文浏览量: 225
- 被引次数: 0