基于高时空分辨率可见光遥感数据的热带山地橡胶林识别

20世纪90年代以来橡胶林种植面积在西双版纳地区迅猛扩大,对该区域橡胶林种植面积、种植结构变化的精确监测是客观评价该地区橡胶林种植与生态环境变化关系的关键。针对西双版纳热带山地地区植被光谱特征的相似性及地形和气候条件的复杂性问题,结合该地区橡胶林冬季落叶的物候特征,采用时空数据融合算法,分别选取中分辨率的ETM+、OLI、Sentinel-2A数据与高时间分辨率的MODIS数据融合,建立高时空分辨率可见光遥感数据集,并分析不同融合数据源对热带山地环境下橡胶林识别精度的差异。结果表明:(1)基于时空融合数据提取的橡胶林物候变化特征能够实现较高精度的橡胶林识别,识别精度可以达到89%以上,Kappa高于0.83;(2)运用10m分辨率的Sentinel-2A数据进行分类时,能够获取比Landsat数据更高精度的分类结果,表明Sentinel-2A数据在高时空数据融合及热带植被遥感应用中有较好前景。     

Google Earth Engine在地球科学与环境科学中的应用研究进展

21世纪以来,随着全球信息化与工业化的高度集成发展,出现了物联网与云计算,人类进入大数据时代。在地学、环境科学及相关学科领域,海量地理、遥感及社会经济等数据产生,在本地平台存储、管理以及分析数据的传统方式已经较难满足当前需求。Google Earth Engine(GEE)云平台由Google云基建提供,是一个对海量地球科学数据集(尤其是遥感影像数据)进行全球尺度在线处理分析和可视化的云计算平台,它利用谷歌强大的计算能力,可以分析处理多种环境与社会问题,如气候变化、植被退化、粮食安全和水资源短缺等。首先对GEE云平台进行介绍,综述了近年来应用GEE云平台所做的相关研究,然后应用该平台及MODIS土地覆盖类型数据,研究了2002~2013年三峡库区主要土地覆盖类型的时空变化规律。结果表明:以林地、灌丛草地以及耕地变化最为明显。最后,经粗略统计得出GEE云平台无论在成本还是效益方面,其综合效率提升90%以上。GEE云平台不仅可以为地学及遥感领域专家提供强有力的支持,也能为相关学科领域人员进行科学研究提供帮助,是一个高效的科研工具。     

基于GEE云平台和Sentinel-2数据的普洱市森林覆盖制图

快速准确获取森林的空间分布对评估森林资源和生态环境状况具有重要的意义。以云南省普洱市为研究区,基于Google Earth Engine(GEE)平台和Sentinel-2影像数据,结合实地调查数据、机载遥感数据及地形辅助数据,提取影像的光谱特征、纹理特征以及地形特征,通过特征筛选,得到适合森林分类的最优特征数据集。结合简单线性非迭代聚类（SimpleNon-Iterative Clustering,SNIC）超像素分割算法,探究不同分类方法、特征变量等因素对分类精度的影响。结果表明：面向对象分类方法的分类精度要优于基于像元分类方法,分类总体精度为88.21%,Kappa系数为0.87,可以较为准确地对普洱市进行森林覆盖制图。面向对象方法可以有效减轻“椒盐现象”,特征优选避免了冗余信息对分类结果的影响,有效提高了分类效率。GEE平台与面向对象方法结合可以提供大区域、高精度的森林覆盖遥感快速制图。     

云南香格里拉地区森林优势树种决策融合分类

基于Google Earth Engine（GEE）云计算平台,协同Sentinel-2影像、WordClim生物气候数据、SRTM地形数据、森林资源二类调查数据等数据,以随机森林（Random Forest, RF）,支持向量机（Support Vector Machine, SVM）和最大熵（Maximum Entropy, MaxEnt）3种机器学习算法为组件分类器,开展多源特征、多分类器决策融合的优势树种分类研究。通过3种组件分类器分别构建了两种串行集成和3种贝叶斯并行集成模型,用于确定云南香格里拉地区10种主要优势树种的空间分布。分类结果显示：3个组件分类器的总体精度均低于67.17%;3种并行集成方法总体精度相当,约为72%;两种串行集成方法精度高于78.48%,其中MaxEnt-SVM串行集成方法获得最佳精度（OA：80.66%, Kappa：0.78）,与组件分类器相比精度至少提高了13.49%。研究表明：决策融合方法在优势树种分类中比组件分类器精度更高,并且有效改善了小样本树种的分类精度,可用于大范围山区优势树种分类。     

基于GEE云平台的黄河源区河流径流量遥感反演研究

河流径流量是陆地上最重要的水文要素之一,准确获取径流信息对于区域的水资源评价和生态修复方面都具有重要作用.研究基于Google Earth Engine(GEE)云平台提供的Sentinel-1、Sentinel-2影像数据,结合数字高程模型(DEM)对河长、河宽、糙率、比降、河深和流速等水力学参数进行遥感估算,进而采...     

基于Sentinel-2影像的多种水体指数法提取地表水对比研究

高时空分辨率的Sentinel-2影像日渐成为地表水体提取的主要遥感数据源,开展基于该卫星影像的多种水体指数方法提取效果的对比研究,对提升地表水遥感监测能力具有重要参考价值。本研究针对目前较为常用的7种水体指数（NDWI、MNDWI、AWEInsh、AWEIsh、WI2015、CDWI和MNDWI＿VIs）,以分布在华北、东北、长江中下游和西北的具有不同地表水体类型组合特征的4个样区为例,在GEE(Google Earth Engine)平台上采用Sentinel-2 MSI影像实现了基于7种水体指数的地表水提取,进而定量分析了不同指数提取水体的精度。结果表明：总体而言,7种水体指数均可以较好识别地表水,但在不同类型的地表水体提取时的表现存在一定的差异;NDWI指数在瞬时性水体（如水田、洪泛区等）会低估地表水的分布,漏分率较高;而AWEInsh、AWEIsh和WI2015指数整体存在高估倾向,错分率较高;MNDWI＿VIs水体指数在复杂水体类型的区域提取精度保持最高;在长时序水体变化监测方面,7种水体的性能表现与基于单景影像所得结论基本一致。本研究为不同类型水体开展地表水监测提供了重要...     

基于Google Earth Engine评估新疆西南部MODIS积雪产品

Google Earth Engine(GEE)是一种基于云建立的地理空间处理平台,可以针对地理空间数据进行分析,实现全球范围内海量遥感数据的并行处理,为遥感大数据、大区域研究提供支持。MODIS积雪覆盖制图是利用MODIS资料建立的全球积雪覆盖产品,已广泛应用于区域乃至全球的气候与环境监测中。GEE云平台存储着百万景遥感影像,其中包括覆盖全球的MODIS逐日积雪产品MOD10A1V5数据和Landsat数据。以新疆西南部3个研究区为例,选取GEE云计算平台存储的Landsat数据,应用NDSI提取积雪范围作为地表覆盖真值,对MOD10A1展开精度评估。结果表明:2000~2016年新疆西南部积雪季MOD10A1的平均总体准确率达82%,平均误判率为2.9%,平均漏判率为58.8%。在晴空条件下,MOD10A1总体准确率可达98%,不同区域的地形及云量是影响MOD10A1精度评估的主要因素。GEE云计算平台可以快速有效地筛选高质量无云的Landsat数据,对全球范围内积雪区的MOD10A1进行精度评估,以在线地图的形式直观显示误判和漏判区域,并利用GEE提供的简单云分函数计算区域云量,使云量对MOD10A1积雪分类精度的影响更具区域代表性。     

Sentinel-2数据的小型湖泊水生植被类群自动提取方法——以翠屏湖为例

针对传统的水生植被遥感监测研究大多是面向大型浅水湖泊,利用Landsat和MODIS数据开展,且很少关注水生植被主要类群的细分,该文以小型湖泊-翠屏湖为例,利用欧空局Sentinel-2高分卫星数据,基于不同水生植被类群及水体间的光谱特征差异,构建了浮叶类植被指数(floating-leaved aquatic vegetation index,FAVI)和沉水植被指数(submerged aquatic vegetation index,SAVI)2个新的植被指数作为分类特征,结合Otsu算法,实现翠屏湖浮叶类植被类群、沉水植被类群和水体的自动提取。经验证,总体分类精度为88.57%,Kappa系数为83.78%,并通过多期影像开展了算法的普适性检验。本研究为快速获取小型浅水湖泊的水生植被类群提供了高效的方法,可为湖泊管理和生态修复提供科学依据。     

Sentinel-2A与Landsat-8影像在油菜识别中的差异性研究

油菜是中国最重要的农作物之一,准确、及时掌握高精度的油菜面积具有重要意义。与Landsat-8数据相比,新一代光学卫星Sentinel-2A数据具有众多优点,但是Sentinel-2A数据在农作物识别方面的应用效果是否一定优于Landsat-8数据仍然是个未知的问题。因此,以油菜最佳识别期内的Sentinel-2A和Landsat-8影像各一景为数据源,选取种植结构复杂的小尺度都市农业区为研究区,基于影像的光谱特征与植被指数信息利用不同分类方法提取油菜种植面积。通过比较不同分类条件、不同方法下的两种影像的油菜识别精度,结果表明:(1)Sentinel-2A影像中不同地物的光谱特征差异与植被指数可分离性高于Landsat-8影像;(2)支持向量机(SVM)分类器下,Sentinel-2A数据的光谱特征获得的油菜制图精度与用户精度最高,分别为89.7%和91.3%,比同等条件下的Landsat-8油菜识别精度分别高7.0%和6.2%;(3)加入纹理信息后,两种数据的总体精度和Kappa系数明显提高,但油菜的制图精度与用户精度并无明显提升。以上结果表明:与Landsat-8数据相比,Sentinel-2A数据能够在种植结构复杂的小尺度区域提取更高精度的作物分布信息。研究结果可以为Sentinel-2A数据的农作物识别与应用提供理论基础。     

基于面向对象的多源卫星遥感影像玉米倒伏面积提取

风灾引起的玉米倒伏可能导致玉米大量减产,利用遥感技术准确监测玉米倒伏面积与空间分布信息对灾情的评估非常重要。利用Planet和Sentinel-2影像分别结合面向对象与基于像元方法提取研究区玉米倒伏,同时评估了不同影像特征（光谱特征、植被指数和纹理特征）与不同分类方法（支持向量机法SVM、随机森林法RF和最大似然法MLC）对玉米倒伏提取精度的影响。结果表明：①使用高空间分辨率的Planet影像进行玉米倒伏提取的精度普遍高于Sentinel-2影像;②从分类精度和面积精度来看,Planet影像的光谱特征+植被指数+均值特征结合面向对象RF分类,总体精度和Kappa系数分别为93.77%和0.87,面积的平均误差最低为4.76%;③采用Planet和Sentinel-2影像结合面向对象分类提取玉米倒伏精度高于基于像元分类。研究不仅分析了面向对象方法的优势,还评估了使用不用影像数据结合面向对象方法的适用性,可以为遥感提取作物倒伏相关研究提供一定的借鉴。     

Research Progress on the Application of Google Earth Engine in Geoscience and Environmental Sciences

With the rapid development and large integration of global informatization and industrialization since the 21st century,the Internet of things and cloud\|computing have emerged.The world has entered an era of big data.There are a huge amount geographical and remote sensing data generated every day in the field of geoscience,environmental science and related disciplines.However,the traditional approaches for storing,managing and analyzing massive data on the local platform,which take up lots of resources,time and energy,have been unable to meet the needs of the current researches.Google Earth Engine(GEE) cloud platform is powered by Google’s cloud infrastructure,and it combines a large number of geospatial datasets and satellite imagery,in which the datasets could be processing,analyzing as well as visualizing on a global scale.Meanwhile,it uses Google’s powerful computational capabilities to analyze and process a variety of environmental and social issues including climate change,vegetation degradation,food security and water resource shortages.Firstly,an introduction of GEE cloud platform has been given.Secondly,recent researches that using GEE cloud platform were reviewed.Thirdly,GEE cloud platform and MODIS land cover type data were used to analyze spatio\|temporal changes patterns of major land use and land cover type in Three Gorges Reservoir in the period of 2002~2013.The results indicate the largest changes occurring in forest lands,shrub grasslands and croplands.Finally,after a rough calculation,GEE cloud platform is superior to the traditional approaches in terms of both cost and economic efficiency,improving the overall efficiency by more than 90%.GEE cloud platform could not only provide powerful support to experts in the field of geosciences and remote sensing,but also offer valuable help to researchers in related disciplines.GEE cloud platform is an excellent tool for scientific research in geosciences,environment sciences and related disciplines.     

Study on Differences between Sentinel-2A and Landsat-8 Images in Rape Identification

Rape is one of the most important crops for many countries,so it is important to obtain accurate rape area.Compared with Landsat-8 data,Sentinel-2A has many advantages,but whether the results of Sentinel-2A data in crop identification are better than Landsat-8 is still an unknown question.The study site is located in a typical agricultural region：Gaochun District in Nanjing,the capital of Jiangsu Province,China,with central coordinates of 118°52′E and 31°19′N.One Sentinel-2A and one Landsat-8 image were obtained during the flowering stage of rape,and then rape area was extracted by using different classification methods based on spectral characteristics and vegetation indices.By comparing the identification accuracy of two images under different classification conditions and methods,the results show that：(1) The difference of spectral characteristics and separability of vegetation indices of different objects in Sentinel-2A were higher than those of Landsat-8 images;(2) Under the classifier of support vector machine,the Producer’s and User’s accuracy of rape of Sentinel-2A based on spectral characteristics were 89.7% and 91.3% respectively,which were 7.0% and 6.2% higher than the identification accuracy of Landsat-8 data;(3) After adding texture information,the overall accuracy and kappa coefficient of two kinds of data were significantly improved,but there was no increase in the producer’s and user’s accuracy of rape.The result presented in this paper show that compared with Landsat-8 data,Sentinel-2A data is more suitable for extracting crop distribution information in small areas with complex planting structure,which can lay a theoretical foundation for crop identification and application of Sentinel-2A data.     

Land Use Intensity and Radiation Feedback in Beijing from 2000 to 2015

Many current studies focus on urban expansion and its heat island effect, but the impact of different land use intensity on radiant energy needs further analysis. Based on the land use data of 2000 and 2015 in Beijing, this study divided the land use of Beijing into five types according to the influence degree of human activities and vegetation resilience, namely, the old urban areas, urban expansion areas, unchanged cropland areas, mixed pixel areas with changed gridcells, and unchanged pure pixel areas. On this basis, we calculated Radiative Forcing (RF) due to the change of surface albedo and explored the relationship between RF and vegetation cover. The results showed that: (1) In pure pixel areas, natural vegetation had a lower albedo, and the corresponding RF was larger than the other four land use type areas. However, under the influence of human activities, RF in the four land use type areas showed an obvious increasing trend during the research period, and the increment was also larger than RF in PP areas. (2) Comparing with unchanged pure pixel, the EVI within the other four human-affected land type areas (old urban areas, urban expansion areas, mixed pixel, and unchanged cropland) decreased but the LOS extended. The combined effect of LOS and EVI contributed to the decreasing trend of surface albedo, which prompted the increase of RF. Our finding highlights that human activities often enhances RF by affecting the intensity of land use. This study has important reference value for analyzing the climate feedback of land use change from physical mechanism.     

2000~2015年北京市土地利用强度及其辐射反馈评估

当前很多研究关注城市扩展及其热岛效应,但不同土地利用强度对辐射能量的影响尚有待进一步的分析。以北京市为例,基于2000和2015年的土地利用数据,按照人为活动对土地的利用程度,将北京市的土地利用变化划分为5类：即老城区、城市扩展区、混合变化区、耕地及自然纯像元区。在此基础上,在反照率和太阳辐射遥感反演数据的支持下,分析2000~2015年由地表反照率引起的辐射强迫(RF, Radiative Forcing),并探讨了RF与植被的关系。结果表明：相比自然纯像元区域,老城区、城市扩展区、混合变化区及耕地的RF在研究时段内均明显增加,后三年的RF均值比前三年增加了0.78 W/m²以上,远大于自然纯像元区的RF增量(0.19 W/m²)。本研究同时发现,植被绿度随土地利用强度的增加而逐年下降,但植被生长期长度却有所延长,两者综合作用于地表反照率,促使了RF的增加,说明单纯从辐射平衡来讲,北京市的土地利用变化在一定程度上增强了RF。     

Evaluation of MODIS Snow Products in Southwestern Xinjiang Using the Google Earth Engine

Google Earth Engine(GEE) is a cloud\|based geospatial processing platform that can analyze geospatial data to achieve parallel processing of massive remote sensing data on a global scale,providing support for remote sensing big data and large\|area research.MODIS snow cover mapping is a global snow cover product established using MODIS data and has been widely used in regional and global climate and environmental monitoring.In the GEE,millions of remote sensing images are stored,including MODIS daily snow products MOD10A1 V5 data and Landsat data.Taking the three research areas in southwestern Xinjiang as examples,the Landsat stored by the GEE were selected,and the NDSI was used to extract the snow cover as the true value of the land cover to evaluate the MOD10A1 accuracy.The results show that the average overall accuracy of MOD10A1 in the snow cover season in southwestern Xinjiang during the period from 2000 to 2016 is 82%,the average misjudgment rate is 2.9%,and the average missed rate is 58.8%.The overall accuracy of MOD10A1 can reach 98% under the clear sky conditions.The accuracy of MOD10A1 is effected by the terrain conditions and cloud cover in different regions.Therefore,the GEE can quickly and effectively filter high quality cloudless Landsat images,and evaluate the accuracy of the MOD10A1 in the snow area around the global regions,displaying intuitively the misjudgment and missed areas in the form of online maps.Meanwhile,GEE provides the Landsat simple cloud score function to calculate the regional cloud cover,which makes the influence of cloud cover on the MOD10A1 accuracy assessment more regionally representative.     

南方地区复杂条件下的耕地面积遥感提取方法

针对我国南方地区植被类型复杂、地形复杂和地块破碎等原因导致耕地信息提取精度较低问题,提出了一种面向对象和CART决策树结合的复杂条件下耕地面积提取方法。以广西南宁市隆安县与武鸣县地区为研究区,采用Sentinel-2A影像,结合数字高程数据（Digital Elevation Model,DEM）及归一化植被指数（Normalized Difference Vegetation Index,NDVI）等多源数据,利用面向对象分割技术识别地块信息,然后以地块为单位采用CART（Classification And Regression Tree,CART）决策树分类法,依据不同地类的形状、光谱特征,提取研究区的耕地。结果表明：面向对象的CART决策树分类方法分类总体精度和Kappa系数分别为96.1%和0.94,相比较于未加入面向对象分割的CART决策树耕地信息提取总体精度提高Kappa系数提高0.54,面向对象的分割方法有利于减少复杂背景对耕地提取的影响。基于面向对象的CART决策树分类方法相比较于传统方法对研究区耕地信息的提取有较好的精确性,能够提高耕地信息的提取精度。     

基于Sentinel-2波段/产品的图像云检测效果对比研究

高时空分辨率遥感影像对精细尺度土地利用和土地覆盖变化研究具有重要意义,然而云噪声的存在给影像的解译和分析带来了一定的挑战,因此云噪声检测作为一项基础性工作在影像解译与分析过程中扮演了非常重要的作用。QA60产品被广泛推荐为Sentinel-2卫星影像的常规云检测产品,然而,我们最近的研究发现基于QA60产品的云检测通常会出现明显的云噪声漏检测现象。为探索提高Sentinel-2卫星影像云噪声检测效果的方法,基于Google Earth Engine（GEE）平台,结合Sentinel-2卫星影像2A级（L2A）数据的2个云相关波段（B1和B9）以及4个产品（QA60、AOT、MSK_CLDPRB和SCL产品）,设计相应分割算法,并以典型区为案例,从影像波段特性、云微物理学等角度分析了相关波段/产品云检测结果的空间分布格局及差异,并借助定量化指标对云检测效果进行评价。结果表明：①在云检测算法方面,B1和B9波段采用的动态阈值分割算法稳健性较好,检测结果能在一定程度上拟合其波段特性,并合理地表征相应波段的云噪声;②从云检测空间分布看,AOT产品效果较差,B9波段和QA60产品云检测可靠性较低,B1、SCL、MSK_CLDPRB 3个波段/产品的云检测潜力较强;③从评价结果看,B1波段的云检测效果最佳,对云噪声的敏感度高于其他云相关波段/产品,查准率、查全率、准确率和F1分数均大于0.90,稳健性最好。本文验证了气溶胶（B1）波段对云检测的精确性、稳健性和敏感程度,有望为进一步优化常规云检测算法提供新参考。     

Cloud and cloud shadow masking for Sentinel-2 using multitemporal images in global area

ABSTRACT

Sentinel-2 data provided the opportunity for complementary data to existing missions including Landsat and SPOT. In this study, multitemporal cloud masking (MCM) used to detect cloud and cloud shadow masking for Landsat 8 was improved to detect cloud and cloud shadow for Sentinel-2 data. This improvement takes advantages of the spectral similarity between Landsat 8 and Sentinel-2. To assess the reliability of the new MCM algorithm, several data selected from different environments such as sub-tropical South, tropical, and sub-tropical North were evaluated. Moreover, these data have heterogeneous land cover and variety of cloud types. In visual assessment, the algorithm can detect cloud and cloud shadow accurately. In the statistical assessment, the user’s and producer’s accuracies of sample in sub-tropical environments of cloud masking was 99% and 96%, respectively, and cloud shadow masking was 99% and 98%, respectively. In addition, the user’s and producer’s accuracies of sample in tropical environments of cloud masking was 100% and 95%, respectively, and cloud shadow masking was 100% and 92%, respectively. Compared to Fmask, MCM has higher accuracies in most of the results of cloud and cloud shadow masking in both sub-tropical and tropical environments. The results showed that the improved-MCM algorithm can detect cloud and cloud shadow for Sentinel-2 data accurately in all scenarios and the accuracies are significantly high.     

Mapping paddy rice fields by applying machine learning algorithms to multi-temporal Sentinel-1A and Landsat data

Sentinel-1A synthetic aperture radar (SAR) data present an opportunity for acquiring crop information without restrictions caused by weather and illumination conditions, at a spatial resolution appropriate for individual rice fields and a temporal resolution sufficient to capture the growth profiles of different crop species. This study investigated the use of multi-temporal Sentinel-1A SAR data and Landsat-derived normalized difference vegetation index (NDVI) data to map the spatial distribution of paddy rice fields across parts of the Sanjiang plain, in northeast China. The satellite sensor data were acquired throughout the rice crop-growing season (May–October). A co-registered set of 10 dual polarization (VH/VV) SAR and NDVI images depicting crop phenological development were used as inputs to Support Vector Machine (SVM) and Random Forest (RF) machine learning classification algorithms in order to map paddy rice fields. The results showed a significant increase in overall classification when the NDVI time-series data were integrated with the various combinations of multi-temporal polarization channels (i.e. VH, VV, and VH/VV). The highest classification accuracies overall (95.2%) and for paddy rice (96.7%) were generated using the RF algorithm applied to combined multi-temporal VH polarization and NDVI data. The SVM classifier was most effective when applied to the dual polarization (i.e. VH and VV) SAR data alone and this generated overall and paddy rice classification accuracies of 91.6% and 82.5%, respectively. The results demonstrate the practicality of implementing RF or SVM machine learning algorithms to produce 10 m spatial resolution maps of paddy rice fields with limited ground data using a combination of multi-temporal SAR and NDVI data, where available, or SAR data alone. The methodological framework developed in this study is apposite for large-scale implementation across China and other major rice-growing regions of the world.