Fronts in Large Marine Ecosystems

Oceanic fronts shape marine ecosystems; therefore front mapping and characterization are among the most important aspects of physical oceanography. Here we report on the first global remote sensing survey of fronts in the Large Marine Ecosystems (LME). This survey is based on a unique frontal data archive assembled at the University of Rhode Island. Thermal fronts were automatically derived with the edge detection algorithm of (Cayula and Cornillon, 1992), (Cayula and Cornillon, 1995) and (Cayula and Cornillon, 1996) from 12 years of twice-daily, global, 9-km resolution satellite sea surface temperature (SST) fields to produce synoptic (nearly instantaneous) frontal maps, and to compute the long-term mean frequency of occurrence of SST fronts and their gradients. These synoptic and long-term maps were used to identify major quasi-stationary fronts and to derive provisional frontal distribution maps for all LMEs. Since SST fronts are typically collocated with fronts in other water properties such as salinity, density and chlorophyll, digital frontal paths from SST frontal maps can be used in studies of physical–biological correlations at fronts. Frontal patterns in several exemplary LMEs are described and compared, including those for: the East and West Bering Sea LMEs, Sea of Okhotsk LME, East China Sea LME, Yellow Sea LME, North Sea LME, East and West Greenland Shelf LMEs, Newfoundland–Labrador Shelf LME, Northeast and Southeast US Continental Shelf LMEs, Gulf of Mexico LME, and Patagonian Shelf LME. Seasonal evolution of frontal patterns in major upwelling zones reveals an order-of-magnitude growth of frontal scales from summer to winter. A classification of LMEs with regard to the origin and physics of their respective dominant fronts is presented. The proposed classification lends itself to comparative studies of frontal ecosystems.     

基于阈值区间的海洋锋面提取模型

A model （Bayesian oceanic front detection, BOFD） of sea surface temperature （SST） front detection in satel- lite-derived SST images based on a threshold interval is presented, to be used in different applications such as climatic and environmental studies or fisheries. The model first computes the SST gradient by using a Sobel algorithm template. On the basis of the gradient value, the threshold interval is determined by a gradi- ent cumulative histogram. According to this threshold interval, front candidates can be acquired and prior probability and likelihood can be calculated. Whether or not the candidates are front points can be deter- mined by using the Bayesian decision theory. The model is evaluated on the Advanced Very High-Resolution Radiometer images of part of the Kuroshio front region. Results are compared with those obtained by using several SST front detection methods proposed in the literature. This comparison shows that the BOFD not only suppresses noise and small-scale fronts, but also retains continuous fronts.     

AMSR-E sea surface temperature (SST) charts enhancement

Microwave satellite images used for retrieving sea surface temperatures often have such distortions as noise and blurring of the thermal fronts. An image processing approach based on the Mumford-Shah model of optimal image approximation is considered for the solution to this problem. We divide images into flat areas and frontal zones, and then process these areas separately. Image fragmentation is based on automatic detection of the thermal front lines. SST enhancement in frontal zones is achieved by using image deconvolution methods. It has been shown that SST errors in high gradient areas reach 1–3 °C. The proposed approach can decrease this discrepancy.     

融合scSE模块的改进Mask R-CNN海洋锋检测方法

海洋锋是重要的中尺度海洋现象,具有数据量小、目标小、弱边缘等特性。针对实际检测任务中弱边缘、小目标海洋锋的检测精度低、错检及漏检率高等问题,融合scSE (spatial and channel Squeeze&Excitation)空间注意力模块构建了一种改进的Mask R-CNN海洋锋检测模型。该方法首先对Mask R-CNN骨干网络结构进行改进,采用scSE模块引导的ResNet-50网络作为特征提取网络,通过加权策略对图像通道和空间位置进行特征突出,提升网络对重要特征的提取能力;其次,针对海洋锋目标边缘定位不准确的问题,引入IoU boundary loss构建新的Mask损失函数,提高边界检测精度。最后,为验证方法的有效性,从训练数据和实验模型上,分别设计多组对比实验。实验结果表明,相比传统Mask R-CNN、YOLOv3神经网络及现有Mask R-CNN改进网络,本文方法对SST梯度影像数据集上的强、弱海洋锋检测效果最好,定位准确率（IoU,Intersection-over-union））及检测精度（mAP,Mean Average Precision）均达0....     

夏季南海北部粤东陆架锋面的动力特征分析*

本文利用现场观测资料和卫星遥感数据, 并结合ROMS(regional ocean modeling system)数值模拟对南海北部粤东陆架的锋面特征及其影响因素进行探讨。观测结果显示, 夏季南海北部陆架存在活跃的上升流温度锋面, 其水平尺度约为50km, 强度达到0.06℃∙km^-1, 大于同时期卫星遥感观测结果, 垂向影响深度超过20m, 且具有一阶理查森数(Richardson number, Ri)的典型动力学特征。进一步的ROMS 模式诊断分析结果显示, 锋面处水平梯度增强, 且动力学上表现出一阶Ri数, 为锋面不稳定的发生提供了有利条件。高分辨率模拟结果显示, 在夏季西南风的驱动下, 沿锋面地转流方向的风应力引起的跨陆架Ekman输运将锋面处冷水向暖水运移, 导致水平浮力梯度和锋面强度增强并形成负Ertel位涡(Ertel potential vorticity, EPV)。因此, 夏季风场强迫引起的Ekman浮力通量(Ekman buoyancy flux, EBF)可能是南海北部锋面不稳定现象的主要贡献者, 对局地动力环境有重要影响。     

我国近海陆架锋面与生态效应研究回顾

海洋锋面存在于特征明显不同的2种或多种水系或水团交界处,锋面区域形成的次生环流和辐聚作用可显著影响到海洋中的物质输运与生物生产,故受到海洋学家的广泛关注。研究发现,我国近海陆架存在14个永久性的准静止锋面（渤海海峡锋面、山东半岛沿岸锋面、苏北沿岸锋面、西韩湾锋面、京畿湾锋面、济州岛西锋面、长江环形浅滩锋面、闽浙沿岸锋面、黑潮锋面、台湾沿岸锋面、闽粤沿岸锋面、珠江口沿岸锋面、琼东锋面和北部湾锋面）,且部分海域观测到双锋面、穿刺锋面和锋面波等现象。它们与陆架环流及其他动力过程（如：涡旋、内波等）共同控制着我国边缘海的物质能量输运与交换以及生物生产力格局。近岸物质沿锋面、跨锋面输运与锋区的垂向输送过程对我国边缘海生物地球化学循环和生态过程存在显著季节性影响。冬季到春季,沿岸锋面松弛能够加强物质从近岸向陆架的输运,进而在空间上调制春季藻华暴发的时间与量级;夏季到秋季,我国边缘海存在显著的潮汐锋面系统,锋面的辐聚效应以及次级环流可显著提高锋面区域的营养盐浓度和改善光照水平,对浮游植物的生长聚集起到促进作用,故在富营养化的河口与沿岸海域,锋面区域容易成为赤潮或缺氧高发区。此外,锋面的物理屏障作用使得两侧水团保持相对独立的物理与化学特征,因而在我国边缘海生境区划和生物多样性梯度变化等方面扮演重要角色,这些研究对认识我国边缘海物质循环与生物生产的控制机制具有重要作用。未来仍需充分结合观测与卫星资料,运用多过程耦合的高分辨率模型,深入认识锋面的精细结构与动态变化,加强亚中尺度和小尺度过程及其生态效应的研究。     

我国周边海域典型海洋锋特征建模研究

基于海洋锋空间位置、水平分布结构和垂直扩展特征等时空特征参数,结合海洋锋空间结构几何模型,建立了区域海洋锋温盐三维结构快速重构特征模型,对黄海西部沿岸锋和东海黑潮中段锋锋区温度场进行了仿真计算,并与实测数据进行了比较分析,实验结果表明：仿真结果与实测数据符合较好,实验结果验证了特征模型的有效性和可推广性。海洋锋区声速具有明显的水平梯度变化,对声纳的水下探测和反探测产生显著影响,因此,需要建立实时估计获取锋区水下温、盐结构的方法。海洋锋特征模型能够快速有效地重构海洋锋区温度场,为实时获取海洋锋水下结构特征提供了方法。     

南海西部风驱离岸急流次中尺度锋面的动力学分析

本文利用卫星观测资料和500 m分辨率数值模拟结果,结合理论分析,对南海西部夏季风场驱动的离岸急流海域次中尺度锋面及其不稳定对背景流场的动力学影响进行了研究。卫星观测和模拟结果表明,南海西部（WSCS）存在侧向尺度为O（1-10）km的次中尺度锋面,在地转和非地转运动的共同作用下,次中尺度密度锋面具有一阶Rossby（Ro）和Richardson（Ri）数。锋面诊断结果显示,沿锋面急流方向的风场强迫引起了显著的跨锋面Ekman净输送,有效地在跨锋面方向将表层冷水平流输送至暖水侧,导致海表浮力损失。减弱的垂向层结和增强的水平浮力梯度使得锋面海域出现负Ertel位涡（PV）,表明该密度锋面易受次中尺度对称不稳定（SI）的影响。次中尺度锋面不稳定引起的跨锋面次级环流能够显著增强垂向速度,其最大值可达100 m·d^-1。能量评估结果表明,次中尺度湍流的两个主要能量源,即地转剪切项（GSP）和垂向浮力通量（BFLUX）在锋面海域显著增强表明在沿锋面急流方向的风场强迫作用下,大尺度地转流的地转剪切动能和锋面有效位能能有效地通过锋面不稳定向次中尺度过程传递。因此,次中尺度锋面及其不稳定有助于增强局地垂向交换和正向串级地转能量,可以为夏季WSCS高叶绿素浓度的相干结构和锋面地转能量的正向传递提供新的动力解释。     

Plankton distribution associated with frontal zones in the vicinity of the Pribilof Islands

We studied the effect of four types of fronts, the coastal front, the middle front, the shelf partition front and the shelf break front on the quantitative distribution and the composition of plankton communities in the Pribilof area of the eastern Bering Sea shelf in late spring and summer of 1993 and 1994. The coastal fronts near St. Paul and St. George Islands and the coastal domains encircled by the fronts featured specific taxonomic composition of planktonic algae, high abundance and production of phytoplankton, as well as large numbers of heterotrophic nanoplankton. The coastal fronts also were characterized by high values of total mesozooplankton biomass, high concentrations of Calanus marshallae, as well as relatively high abundances of Parasagitta setosa and Euphausiacea compared to surrounding shelf waters. We hypothesize that wind-induced erosion of a weak thermocline in the inner part of the coastal front as well as transfrontal water exchange in subthermocline layers result in nutrient enrichment of the euphotic layer in the coastal fronts and coastal domains in summer time. This leads to prolonged high primary production and high phytoplankton biomass. In this paper a new type of front—the shelf partition front located 45–55 km to the north-east off St. Paul Island—is described, which is assumed to be formed by the flux of oceanic domain waters onto the shelf. This front features a high abundance of phytoplankton and a high level of primary production compared to the adjacent middle shelf. Near the southwestern periphery of the front a mesozooplankton peak occurred, composed of C. marshallae, with biomass in the subthermocline layer, reaching values typical for the shelf break front and the highest for the area. High abundance of phyto- and zooplankton as well as heterotrophic nanoplankton and elevated primary production were most often observed in the area adjacent to the shelf break front at its oceanic side. The phyto- and mesozooplankton peaks here were formed by oceanic community species. The summer levels of phytoplankton numbers, biomass and primary production in the shelf break frontal area were similar to those reported for the outer and middle shelf during the spring bloom and the coastal domains and coastal fronts in summer. In the environment with a narrow shelf to the south of St. George Island, the mesozooplankton peak was observed at the inner side of the shelf break front as close as 20 km from the island shore and was comprised of a “mixed” community of shelf and oceanic species. The biomass in the peak reached the highest values for the Pribilof area at 2.5 g mean wet weight m⁻³ in the 0–100 m layer. Details of the taxonomic composition and the numbers and production of phytoplankton hint at the similarity of processes that affect the phytoplankton summer community in the coastal domains of the islands, at the coastal fronts, and at the oceanic side of the shelf break front. The middle front was the only one that had no effect on plankton composition or its quantitative characteristics in June and July. Location of a variety of frontal productive areas within 100 km of the Pribilof Islands creates favorable foraging habitat for higher trophic level organisms, including sea birds and marine mammals, populating the islands.     

基于海色融合数据的东海黑潮锋面检测研究

海洋锋是典型的海洋中尺度现象之一。目前卫星遥感主要利用海表温度数据分析海洋锋,但由于西北太平洋海域夏季海表温度的趋同特性,不能进行有效的锋面监测;而不同水团所具有的生物光学特性往往是不同的,且不具有太阳辐射引起的显著性季节变化,因此海色资料也成为检测海洋锋的有效数据源。文中以东海黑潮为例,详细说明了基于叶绿素a浓度融合数据,采用梯度法进行海洋锋面检测的过程,通过比较不同季节不同梯度阈值得到的东海黑潮锋结果,从保持锋面的完整性及对零碎锋区的剔除效应方面,选取了不同季节较优的梯度阈值。总体来说,文中检测出的东海黑潮区域海色锋与海流黑潮强流区较吻合,12月至4月东海黑潮海色锋检测结果不如海温锋,而5-11月东海黑潮海色锋检测结果优于海温锋,特别是台湾以东黑潮区域,不论什么季节海温锋都没有体现,而海色锋始终很明显。利用文中提出的海洋锋检测算法、分析方法及选择的梯度阈值可以有效地检测东海黑潮区域的海洋锋面,结合海色锋和海温锋,可以监测分析东海黑潮强流区的时空变化。     

基于小波分析的海洋锋形态特征提取

遥感影像中海洋锋具有弱边缘(Weak Edge)信息特性,其空间尺度在不同的空间位置存在较大的差异。针对传统边缘算子只能提取某一尺度的海洋锋信息,而损失其他尺度信息不足,本文利用小波多尺度分析特性,探讨了海洋锋信息在各尺度间的分布的继承性,提出最佳尺度确定的算法框架,设计出最佳尺度确定算法的流程,进一步进行海洋锋各尺度信息的提取。最后以海面温度锋-墨西哥湾流为实例进行形态特征提取,并与典型的边缘提取算子效果在信息量、抗干扰性、定位精度、连续性和对比度上进行对比分析。试验结果表明了该算法在提取多尺度的海洋锋信息上的可行性和实用性。     

基于自动探测方法的南海北部涡致海表温度锋面的特征研究

南海北部具有丰富的温度锋面和中尺度涡,它们调节着局地的热量和能量平衡。本文利用卫星海洋高度异常和海表温度数据,并基于自动探测方法,探究了2007年至2017年南海北部中尺度涡边缘的海表温度锋面（涡致锋面）特征。反气旋/气旋边缘出现锋面的概率可达20%。气旋涡在各个方向上出现锋面的概率比较均匀,反气旋涡的东北部和西南部出现锋面的概率大于西北部和东南部。中尺度涡致锋面的数量有明显的季节变化,而涡动能未表现出明显的季节变化。中尺度涡致锋区的总涡动能是中尺度涡内动能的3倍,并且反气旋涡致锋面的总涡动能明显强于气旋涡致锋面的总涡动能。中尺度涡致锋面的数量和涡动能的年际变化与厄尔尼诺南方涛动指数没有明显的相关性。本研究也讨论了中尺度涡致锋面的可能机制,但是中尺度涡对海表温度锋的贡献需要进一步定量研究。     

Identification of the fronts from the Kuroshio Extension to the Subarctic Current using absolute dynamic topographies in satellite altimetry products

The climatologies and variabilities of the fronts associated with the Kuroshio Extension (KE), the Kuroshio Extension Northern Branch (KENB), the flow along the Subarctic Boundary (SAB), and the Subarctic Current (SAC) are identified by a new method based on the absolute dynamic height product from the archiving, validation, and interpretation of satellite oceanographic data (the AVISO product). The fronts are detected by examining the specific contour values of the absolute dynamic topography (ADT) in the AVISO product. The (time-varying) specific contour values are decided from the local maxima of the averaged surface geostrophic velocity along the ADT. Assuming the specific contours as the front locations, we obtain an occurrence frequency map of the four front locations, and determine the monthly variability of the fronts over the 1993–2015 period. The results are validated by hydrographic observations. The KE and KENB east of the Shatsky Rise migrate southward several times at a speed of ~ 0.2 cm s^?1, while the SAB and SAC are mostly stationary.     

2000—2017年东印度洋季风带海域温度锋时空特征变化分析

基于2000-2017年的MODIS-Terra气候态月平均海表温度数据检测了东印度洋季风带海域的温度锋,统计了各锋面每月发生的锋点数量、锋面平均强度及中心线长度,并基于2000—2017年逐年海表温度数据,研究了以恒河-雅鲁藏布江河口锋和爪哇岛锋为代表的典型温度锋面的年际变化。结果发现:在3、4月,东印度洋季风带海域的锋面最弱、数量最少;5-10月期间,5°~15°N及5°~15°S一带的温度锋出现并发展;12、1、2月最北部锋面发展并趋向成熟。研究区存在恒河-雅鲁藏布江河口锋、爪哇岛锋、Palk海峡锋、东锡兰锋及伊洛瓦底江河口锋5个温度锋,其中最北部的恒河-雅鲁藏布江河口锋全年存在,东北季风时期长度较长,强度较大,最南部的爪哇岛锋存在于4—11月,西南季风时期较强,长度和数量也处于较高水平,其余锋面主要发生于西南季风盛行时期,且强度、长度等变化相对较小。两个典型温度锋空间位置的年际变化均不大,恒河-雅鲁藏布江河口锋平均强度的年际变化较大,最大超过0.03℃/km,长度变化相对较小;爪哇岛锋平均强度的变化相对平稳,但锋点数量和中心线长度存在较大的年际变化。     

卫星视频舰船目标检测方法

舰船目标检测是进行海上目标监管,保障海上权益的重要手段.本文在SSD(single shot multibox detector)算法的基础上,利用残差网络(ResNet,residual network)作为骨干网络构建SSD模型,将改进后的SSD算法应用于卫星视频舰船目标检测,该算法采用残差连接替换原本的级联方式,...     

Frequent locations of oceanic fronts as an indicator of pelagic diversity: Application to marine protected areas and renewables

Frequent locations of thermal fronts in UK shelf seas were identified using an archive of 30,000 satellite images acquired between 1999 and 2008, and applied as a proxy for pelagic diversity in the designation of Marine Protected Areas (MPAs). Networks of MPAs are required for conservation of critical marine habitats within Europe, and there are similar initiatives worldwide. Many pelagic biodiversity hotspots are related to fronts, for example cetaceans and basking sharks around the Isle of Man, Hebrides and Cornwall, and hence remote sensing can address this policy need in regions with insufficient species distribution data. This is the first study of UK Continental Shelf front locations to use a 10-year archive of full-resolution (1.1 km) AVHRR data, revealing new aspects of their spatial and seasonal variability. Frontal locations determined at sea or predicted by ocean models agreed closely with the new frequent front maps, which also identified many additional frontal zones. These front maps were among the most widely used datasets in the recommendation of UK MPAs, and would be applicable to other geographic regions and to other policy drivers such as facilitating the deployment of offshore renewable energy devices with minimal environmental impact.     

The frontal system at the boundary of the East China Sea: Its variability and response to large-scale atmospheric forcing

High-resolution satellite sea surface temperature measurements (PATHFINDER dataset) indicate that the fronts at the boundary of the East China Sea (Taiwan front, Kuroshio frontal zone, and South Korean coastal front) appear as a unified dominating frontal structure when climatological averaging is applied. This structure is about 1200 km in length, spreads over the continental shelf from Taiwan to the Tsushima Islands, and separates productive seawaters from the oligotrophic oceanic waters. The Kuroshio frontal zone, incorporated into this structure, reveals interannual variability with periods consistent with El Niño–Southern Oscillation (4–5 years).     

Detection of meteorological fronts over the North Sea with Valensole skywave radar

The skywave radar located at Valeusole is described. The front localization method is reviewed and discussed. Wind direction was mapped over the North Sea during a three-day experiment. Thirty maps were collected and two cold fronts were successfully detected and tracked. Data corresponding to one front are presented, discussed, and compared to French Meteorological Office (FMO) surface charts. The measurements show that this front did not move with a constant velocity. Front detection is finally discussed in terms of spatial localization accuracy, resolution, and detectability.     

Wavelet analysis of satellite images for coastal watch

The two-dimensional wavelet transform is a very efficient bandpass filter, which can be used to separate various scales of processes and show their relative phase/location. In this paper, algorithms and techniques for automated detection and tracking of mesoscale features from satellite imagery employing wavelet analysis are developed. The wavelet transform has been applied to satellite images, such as those from synthetic aperture radar (SAR), advanced very-high-resolution radiometer (AVHRR), and coastal zone color scanner (CZCS) for feature extraction. The evolution of mesoscale features such as oil slicks, fronts, eddies, and ship wakes can be tracked by the wavelet analysis using satellite data from repeating paths. Several examples of the wavelet analysis applied to various satellite images demonstrate the feasibility of this technique for coastal monitoring