北极海冰密集度动态系点值ASI反演算法研究

海冰密集度是极区海冰监测的重要因素,使用AMSR-E (Advanced Microwave Scanning Radiometer for Earth Observing System) 89GHz数据ASI反演算法得到的海冰密集度是目前能够获得的分辨率最高的微波数据.在以前的算法中往往使用固定的系点值,本研究实现了动态系点值ASI (the Arctic Radiation And Turbulence Interaction Study (ARTIST) Sea Ice)算法,更重要的是在统计开阔水系点值的时候消除了云对系点值的影响,使得纯水系点值更接近真实状况.得到2010年平均的开阔水和海冰的系点值分别为50.8K和7.8K,通过每天的系点值得到的反演方程在低密集度区增大了海冰密集度,在高密集冰区减小了海冰密集度,从而在一定程度上改善了微波数据的反演准确度.通过和北极区域选取40幅不受云影响的MODIS 500m分辨率宽频大气层顶反照率(broadband TOA albedo)计算的海冰密集度进行了比较验证.结果显示,40个个例中,95%本文的平均差异比使用固定系点值算法产品的小,而且75%的均方根差异比使用固定系点值算法产品的小.     

海冰密集度卫星遥感反演研究进展

海冰密集度是海冰的重要参数之一,在冰区导航、海上作业、海冰模式验证和气候模型改进等方面具有重要意义。卫星遥感具有覆盖范围广、重访周期短、成本相对低等优势,已成为获取海冰密集度的主要观测手段。本文从主被动微波遥感和光学遥感的角度,回顾了现阶段海冰密集度卫星遥感反演研究进展情况,包括海冰监测传感器、海冰密集度反演算法和海冰密集度产品等。结果表明,被动微波遥感是目前获取海冰密集度的主要方式,已发展出许多成熟的业务化算法;主动微波遥感数据已成为制作冰情图的主要数据源,海冰密集度反演算法由合成孔径雷达SAR（Synthetic Aperture Radar）图像分类向深度学习算法发展;光学遥感海冰密集度算法较为成熟,但受限于云层和夜晚限制,其反演结果多用于其他海冰密集度产品的验证。受传感器硬件限制,3种观测手段各有其长处与不足。为获得高精度、高时空分辨率的海冰密集度数据,开展多源数据融合研究是解决传感器性能瓶颈的有效手段。大数据时代,基于深度学习的海冰密集度卫星遥感反演技术快速发展,需要深度融入海冰密集度卫星遥感领域知识。海冰密集度卫星遥感反演应着力于海冰预报服务,致力于提高我国的海冰预报能力。     

两种准实时遥感海冰密集度产品在中国第五次北极考察期间的适用性评估

比较了AMSR2和SSMIS产品在2012年中国第五次北极考察期间的差异,并利用雪龙船在北极走航观测的海冰密集度资料初步评估了两种卫星产品在北极东北航道和高纬航道的适用性。结果表明:两种产品在海冰边缘区域反演的海冰密集度差异较大,且在高纬度区域AMSR2反演的密集度普遍大于SSMIS;两种产品对海冰外缘线的反演基本相同,说明两种算法对海冰和海水的区分基本一致;在去程低纬航线上分辨率较高的AMSR2数据的平均偏差为0.14±0.11,而分辨率较低的SSMIS数据为0.17±0.11;在回程高纬航线上AMSR2数据的平均偏差为0.11±0.10,而SSMIS数据为0.11±0.12。SSMIS数据在高值区明显的低估了海冰密集度值,说明其在高值区的反演上存在系统性偏差,AMSR2数据和走航观测数据更相符。SSMIS数据在高值区偏差大的原因可能与其反演算法对海冰表面出现的大量融池的辨别能力较差有关。     

渤海AVHRR多通道海冰密集度反演算法试验研究

为了得到更精确的渤海海冰密集度反演参数,采用辽东湾不同类型海冰ASD实测数据,在分析光谱特征的基础上,针对NOAA/AVHRR数据确定出合适海冰密集度反演算法阈值。继而,基于线性光谱混合模型的多通道反演算法进行了一系列算法试验。同时实现了基于LandSat5-TM数据的渤海海冰密集度场反演,并利用该结果与AVHRR单通道和多通道算法得到的海冰密集度反演结果进行比对分析。定量误差分析结果表明,当单通道算法或组合算法中包含1通道时,与Landsat5-TM反演结果的平均误差为正值,包含2通道且不包含1通道时,平均误差为负值;同时使用这两个通道较只包含其一的各种组合算法的平均误差明显偏小;在各种组合算法中,1245四个通道组合反演的海冰密集度结果误差最小,可应用于渤海AVHRR数据海冰密集度反演。     

基于FY-3C微波辐射计数据的极区海冰密集度反演方法研究

极区海冰影响大气和海洋环流,对全球气候变化起着重要的作用。海冰密集度是表征海冰时空变化特征的重要参数之一。本文研究了利用FY-3C微波扫描辐射计亮温数据反演极区海冰密集度的方法。经过时空匹配、线性回归,修正了FY-3C微波辐射计亮温数据。使用两种天气滤波器和海冰掩模滤除了大气影响所造成的开阔海域虚假海冰;使用最小密集度模板去除陆地污染效应。通过计算2016年、2017年极区海冰面积及范围两个参数,对得到的海冰密集度产品进行了验证,两年的海冰范围和面积趋势基本与NSIDC产品一致,平均差异小于3%。本研究结果为发布我国自主卫星的极区海冰密集度业务化产品奠定了基础,制作的产品可保障面临中断的40多年极区海冰记录的连续性。     

夏季北极密集冰区范围确定及其时空变化研究

研究夏季北极密集冰区的范围变化是了解北极海冰融化过程的重要手段。密集冰区与海冰边缘区之间没有明确的分界线, 海冰密集度在两者之间平滑过渡, 确定密集冰区范围就需确定一个密集度阈值。文中依据分辨率为6.25 km的AMSR-E遥感数据, 发现不同密集度阈值所围范围在密集冰区边缘处的减小存在由快变慢的过程, 同时与周围格点的密集度差异变化在该处最为显著, 对这两个特征进行统计分析, 获得的阈值同为89%, 具有明确的物理意义和合理性。以此为基础, 运用腐蚀算法剔除海冰边缘区, 同时结合连通域法排除小范围密集冰的影响, 进而确定密集冰区的范围。结果表明, 2002-2006年密集冰区覆盖范围较大, 年际变化较小, 2007年以后明显减小, 2010年与2011年相继出现最小值, 其中2011年的范围最小值仅为2006年的64%。密集冰区范围的变化不同于海冰覆盖范围, 是具有独立特性的海冰变化参数, 反映出高密集度海冰区域的变化特征。海冰的融化与海冰边缘区的变化是导致密集冰区范围发生变化的两个主要因素, 受动力学因素的影响, 海冰边缘区发生伸展或收缩, 发生密集冰区与海冰边缘区互相转化。本文提出了一种研究北极海冰变化的新思路, 密集冰区覆盖范围的减小表明北极中央区域高密集度海冰正持续减少。     

基于SMAP卫星雷达资料的海冰密集度反演技术研究

SMAP是美国于2015年初发射的一颗卫星,搭载了L波段的雷达。它采用圆锥扫描方式,具有固定的入射角、较大的幅宽和千米级的分辨率,在海冰监测方面具有独特的优势。本文利用SMAP卫星雷达资料分别与德国Bremen大学海冰密集度产品和美国国家冰雪数据中心（NSIDC）海冰密集度产品建立3.125 km和25 km匹配数据集,分析了L波段雷达后向散射系数、极化比和归一化极化差与海冰密集度之间相关性,建立基于人工神经网络的海冰密集度反演算法。为了验证SMAP卫星雷达资料反演海冰密集度的精度,本文选择德国Bremen大学和美国冰雪数据中心发布的海冰密集度产品分别与SMAP海冰密集度产品进行对比分析,SMAP海冰密集度与Bremen海冰密集度的偏差为0.07、均方根误差为0.14;与NSIDC海冰密集度的偏差为0.04、均方根误差为0.18,这表明SMAP海冰密集度产品与现有业务化海冰密集度产品具有很好的一致性。     

HY-2卫星反演海冰密集度在北极地区的适用性评估

评估了我国自主研发的海洋二号卫星(HY-2)海冰密集度产品在北极地区的适用性。与8种国际同类产品相比,HY-2产品的分辨率为25 km,属于低分辨率产品。HY-2产品2012年夏季的空间分布特征和其他产品差别不大,但在低密集度冰区和边缘区域的差异可达0.15～0.25。HY-2产品可以反映2012年7—10月海冰面积先减小后增大的规律,但最小海冰范围的出现时间比其他产品偏早,且平均值偏小。利用北极海冰数值预报系统进行的同化试验显示,HY-2产品可以有效改善海冰密集度的模拟结果,将平均偏差从控制试验的0.18～0.24减小为同化试验的0.05～0.08,改善效果和国际认可和常用的AMSR2/ASI产品相当。     

基于CryoSat-2卫星测高数据的北极海冰体积估算方法

近30年来,北极海冰正发生着剧烈的变化。海冰体积是量化海冰变化的重要指标之一。本文以2015年CryoSat-2卫星测高数据和OSI SAF海冰类型产品为基础。提取了浮冰出水高度、积雪深度、海冰密集度、海冰类型等属性信息,通过数据内插、投影变换、栅格转换、空间重采样等工作将海冰属性信息统一为25 km×25 km分辨率的栅格数据集。根据流体静力学平衡原理,逐个估算栅格像元对应的海冰厚度值,将其与对应的海冰面积相乘,估算了北极海冰密集度大于75%海域的海冰体积,并分析了海冰厚度和体积的月变化和季节变化特征。用NASA IceBridge海冰厚度产品对反演的海冰厚度进行验证。结果表明二者相关系数为0.72,有较高的一致性。北极海冰平均厚度春季最大,夏季最小,分别约为2.99 m和1.77 m,最厚的海冰集中在格陵兰沿岸北部和埃尔斯米尔半岛以北海域。多年冰平均厚度大于一年冰。冬季海冰体积最大,约为23.30×103 km3,经过夏季的融化,减少了近70%。一年冰体积季节波动较大,而多年冰体积相对稳定,季节变化不明显。     

长序列北极海冰覆盖数据集对比分析

国家卫星气象中心使用2011年至今的风云三号卫星数据开发了一套基于Nasa Team2（NT2）算法的北极海冰密集度数据集,并可实时业务更新。将该数据集与其他国家不同机构业务运行并实时更新多种同类型数据集进行横向对比分析,其中包括:（1）美国冰雪中心基于Nasa Team（NT）算法以及SSM/I、SSMIS数据制作的1978年至今25 km分辨率全球极区海冰覆盖数据集;（2）美国冰雪中心基于Boot Strap（BS）算法以及SSM/I、SSMIS数据制作的1978年至今25 km分辨率全球极区海冰覆盖数据集;（3）美国NOAA基于多种卫星资料、地面观测数据以及海冰模型制作的2004年至今4 km分辨率北半球海冰覆盖数据集（IMS）。对比表明,上述数据集在北极地区不同的时空范围内存在一定的偏差。以分辨率较高的IMS数据集为基准,对其他3种长序列数据集进行初步评价,总体最大偏差超过100×104 km2,其中,NT2数据集过估较明显。经过与IMS数据集多年各月监测最大值的对比订正,NT2数据集过估情况得到改善。在此基础上的分析结果表明,NT、BS、NT2等3种数据集与IMS数据集相比,过估区域主要分布在海岸线附近,夏季过估比冬季更加明显,少估区域与算法、月份相关性明显,夏季少估面积也较冬季更大。NT、BS、NT2等3种数据集之中,NT2数据集与IMS数据集偏差最小,NT数据集次之,BS数据集与IMS数据集偏差最大。结果表明使用风云三号卫星数据的北极海冰覆盖数据集精度与国外3种同类型数据集相当。     

Retrieving the antarctic sea-ice concentration based on AMSR-E 89 GHz data

Sea-ice concentration is a key item in global climate change research.Recent progress in remotely sensed sea-ice concentration product has been stimulated by the use of a new sensor,advanced microwave scanning radiometer for EOS(AMSR-E),which offers a spatial resolution of 6 km×4 km at 89GHz.A new inversion algorithm named LASI(linear ASI) using AMSR-E 89GHz data was proposed and applied in the antarctic sea areas.And then comparisons between the LASI ice concentration products and those retrieved by the other two standard algorithms,ASI(arctic radiation and turbulence interaction study sea-ice algorithm) and bootstrap,were made.Both the spatial and temporal variability patterns of ice concentration differences,LASI minus ASI and LASI minus bootstrap,were investigated.Comparative data suggest a high result consistency,especially between LASI and ASI.On the other hand,in order to estimate the LASI ice concentration errors introduced by the tie-points uncertainties,a sensitivity analysis was carried out.Additionally an LASI algorithm error estimation based on the field measurements was also completed.The errors suggest that the moderate to high ice concentration areas(>70%) are less affected(never exceeding 10%) than those in the low ice concentration.LASI and ASI consume 75 and 112 s respectively when processing the same AMSR-E time series thourghout the year 2010.To conclude,by using the LASI algorithm,not only the seaice concentration can be retrieved with at least an equal quality as that of the two extensively demonstrated operational algorithms,ASI and bootstrap,but also in a more efficient way than ASI.     

基于AMSR-E数据的多年冰密集度反演算法研究

In recent years, the rapid decline of Arctic sea ice area(SIA) and sea ice extent(SIE), especially for the multiyear(MY) ice, has led to significant effect on climate change. The accurate retrieval of MY ice concentration retrieval is very important and challenging to understand the ongoing changes. Three MY ice concentration retrieval algorithms were systematically evaluated. A similar total ice concentration was yielded by these algorithms, while the retrieved MY sea ice concentrations differs from each other. The MY SIA derived from NASA TEAM algorithm is relatively stable. Other two algorithms created seasonal fluctuations of MY SIA, particularly in autumn and winter. In this paper, we proposed an ice concentration retrieval algorithm, which developed the NASA TEAM algorithm by adding to use AMSR-E 6.9 GHz brightness temperature data and sea ice concentration using 89.0GHz data. Comparison with the reference MY SIA from reference MY ice, indicates that the mean difference and root mean square(rms) difference of MY SIA derived from the algorithm of this study are 0.65×106 km2 and0.69×106 km2 during January to March, –0.06×106 km2 and 0.14×106 km2 during September to December respectively. Comparison with MY SIE obtained from weekly ice age data provided by University of Colorado show that, the mean difference and rms difference are 0.69×106 km2 and 0.84×106 km2, respectively. The developed algorithm proposed in this study has smaller difference compared with the reference MY ice and MY SIE from ice age data than the Wang's, Lomax' and NASA TEAM algorithms.     

北极地区不同冰龄的海冰厚度变化研究

In this study, changes in Arctic sea ice thickness for each ice age category were examined based on satellite observations and modelled results. Interannual changes obtained from Ice, Cloud, and Land Elevation Satellite(ICESat)-based results show a thickness reduction over perennial sea ice(ice that survives at least one melt season with an age of no less than 2 year) up to approximately 0.5–1.0 m and 0.6–0.8 m(depending on ice age) during the investigated winter and autumn ICESat periods, respectively. Pan-Arctic Ice Ocean Modeling and Assimilation System(PIOMAS)-based results provide a view of a continued thickness reduction over the past four decades. Compared to 1980 s, there is a clear thickness drop of roughly 0.50 m in 2010 s for perennial ice. This overall decrease in sea ice thickness can be in part attributed to the amplified warming climate in north latitudes. Besides, we figure out that strongly anomalous southerly summer surface winds may play an important role in prompting the thickness decline in perennial ice zone through transporting heat deposited in open water(primarily via albedo feedback) in Eurasian sector deep into a broader sea ice regime in central Arctic Ocean. This heat source is responsible for enhanced ice bottom melting, leading to further reduction in ice thickness.     

Comparative analysis of satellite and shipborne data on ice cover in the Russian Arctic seas

The SSM/I data processed by the NASA Team algorithm are used to compare the total ice concentration obtained from the visual shipborne observations with satellite images. A comparison of the satellite images with the shipborne data shows significant differences with the shipborne data observed onboard ice-breakers during 15 scientific expeditions to the Barents, Kara, Laptev, and East Siberian seas. The most pronounced differences are observed in the ice edge regions. They cause errors in the estimates of the total ice concentration and ice extent, which makes difficult to use them in various practical and scientific tasks. Generally, the methods of remote sensing (RS) underestimate the real sea ice concentration: the average error is on the order of 10% in both winter and in summer. A statistical analysis of the comparison of two sources of information was carried out separately for the total ice concentration for summer and winter data taking into account the new ice and without it. During the summer period in the area of open ice, the SSM/I data over-estimate the total ice concentration by 0.5–1, but in close ice they underestimate it on average by 2 grades. If the new ice is subtracted from the total ice concentration obtained onboard the icebreakers, the total error decreases to ?3.4%. In the winter period in the region of rare ice, the SSM/I data overestimate the total ice concentration by 1–2 grades, but in close ice this difference is as high as 2, like in summer. New ice in winter is determined better by remote sensing methods than in summer; hence, its exclusion from the total ice concentration does not lead to a decrease in the average error.     

Estimation of Sea Ice Thickness from SARAL/AltiKa in Drifting Orbit Phase

Abstract

Intra and inter-annual variations in the sea ice thickness are highly sensitive indicators of climatic variations undergoing in the earth’s atmosphere and oceans. This paper describes the method of estimating sea ice thickness using radar waveforms data acquired by SARAL/Altika mission during its drifting orbit phase from July 2016 onwards yielding spatially dense data coverage. Based on statistical analysis of return echoes, classification of the surface has been carried out in three different types, viz. floe, lead and mixed. Time delay correction methods were suitably selected and implemented to make corrections in altimetric range measurements and thereby freeboard. By assuming hydrostatic equilibrium, freeboard data were converted into sea ice thickness. Results show that sea ice thickness varies from 4 to 5?m near ice shelves and 1 to 2.5?m in the marginal sea ice regions. Freeboard and sea ice thickness estimates were also validated using NASA’s Operation Ice Bridge (OIB) datasets. Freeboard measurements show very high correlation (0.97) having RMSE of 0.13. Overestimation of approximately 1–2?m observed in the sea ice thickness, which could be attributed to distance between AltiKa footprint and OIB locations. Moreover, sensitivity analysis shows that snow depth and snow density over sea ice play crucial role in the estimation of sea ice thickness.     

海洋二号微波散射计极地海冰识别算法研究

An Antarctic sea ice identification algorithm on the HY-2A scatterometer(HSCAT) employs backscattering coefficient(σ0) and active polarization ratio(APR) for a preliminary sea ice identification.Then standard deviation(STD) filtering and space filtering are carried out.Finally,it is used to identify sea ice.A process uses a σ0,STD threshold and an APR as sea ice indicators.The sea ice identification results are verified using the sea ice distribution data of the ASMR2 released by the National Snow and Ice Data Center as a reference.The results show very good consistence of sea ice development trends,seasonal changes,area distribution,and sea ice edge distribution of the sea ice identification results obtained by this algorithm relative to the ASMR2 sea ice results.The accuracy of a sea ice coverage is 90.8% versus the ASMR2 sea ice results.This indicates that this algorithm is reliable.