, 张镱锂, 刘林山Accuracy evaluation of the seven land cover data in Qiangtang Plateau
LIUQionghuan, ZHANGYili, LIULinshan通讯作者:
收稿日期:2017-04-24
修回日期:2017-09-8
网络出版日期:2017-11-20
版权声明:2017《地理研究》编辑部《地理研究》编辑部
基金资助:
作者简介:
-->
展开
摘要
关键词：
Abstract
Keywords：
-->0
PDF (20104KB)元数据多维度评价相关文章收藏文章
本文引用格式导出EndNoteRisBibtex收藏本文-->
1 引言
土地覆被数据是环境、生态等众多科学研究的基础,尤其在土地资源可持续利用以及政府决策中发挥重要作用[1]。随着遥感和GIS技术的发展,涌现出大量以遥感影像为数据源的全球或区域范围的土地覆被产品。截至2016年,有20余套全球尺度及区域尺度免费共享土地覆被数据产品[2]。然而,由于分类系统、分类技术、影像获取时间和空间分辨率等不一致,不同数据应用于区域或者全球尺度时,其精度存在明显的差异[3,4],严重影响土地覆被数据在各个领域的有效利用[5,6]。因此,对已有土地覆被数据资料的精度进行分析非常必要[7-9]。目前,部分****采用相对评价方法,对比多种土地覆被数据在整体范围的面积和空间一致性及差异性[4,10]。如对于不同分辨率产品进行全球或区域土地覆被数据空间一致性比较[11,12]。也有****以某套数据产品为参考,如GLC2000数据,对比评估其他全球土地覆被数据总体精度[13]。然而,大多****以样本点为参考数据,评价和验证数套全球土地覆被数据精度[3,14-18]。因该方法可靠性高,在土地覆被数据精度评价中最为常见,在该方法中,样本点的选取、布设和大小是影响精度评价的关键因素[17]。
近年来,青藏高原的草地退化、冰川退缩、湖泊扩张等生态环境变化越来越受到科研工作者的关注[19-22]。作为青藏高原腹地,羌塘高原的土地覆被变化对自身区域内生态环境起着关键作用,是青藏高原生态环境变化研究的重要组成部分[23]。然而,由于羌塘高原自然环境极其恶劣,在该区域的科学考察较少,研究基础资料匮乏。已有的数值模拟和遥感反演数据的精度也大多缺乏地面观测的验证[23,24]。
本文基于横跨羌塘高原的6851个土地覆被样本点,从总体精度、用户精度和制图精度三个方面,对GlobCover 2009（GlobCover）、University of Maryland（UMD）、和IGBPDIS等七套大尺度土地覆被数据在羌塘高原的精度进行评估,并试图回答：在羌塘高原地区,所选的七套代表性土地覆被数据产品中,数据综合精度如何、哪套数据精度最高?对于主要的土地覆被类型,哪套数据最为合理?影响羌塘高原土地覆被数据精度的原因有哪些,如何改进?
2 研究方法与数据来源
2.1 土地覆被数据产品
已有的20余套全球土地覆被数据产品分别属于11个数据系列[2]。本文选用其中的七套。其中,有5个数据系列——GlobeLand30、CCI-LC、GlobCover、MODIS Land Cover和Global Land Cover 250 m China（GLC250 m_CN）——包含不同年份的数据共22套。有一个数据系列未共享（GLC250 m_CN）,故本文选用了其他4个数据系列（表1）中最新年份的数据（截至2016年12月公开发布）;文中选用的另外3套数据（表1）分别属于3个数据系列（UMD、IGBP DISCover和GLC 2000）,且在长时间序列的土地覆被变化研究中运用广泛。有2个数据系列是同一年份、不同空间分辨率,即Finer Resolution Observation and Monitoring-Global Land Cover（FROM-GLC）和International Satellite Land Surface Climatology Project（ISLSCP II）,该系列共12套数据,本文未选用;另1个数据系列（GeoWiki）,因其年份不确定,亦未选用。Tab.1
表1
表1七套土地覆被数据
Tab.1Characteristics of seven land cover data sets
| 数据名称 | 整体精度（%） | 验证方法 | 传感器 | 分类方法 | 分辨率 | 年份 | 分类系统类型数 | 链接 | 参考文献 |
|---|---|---|---|---|---|---|---|---|---|
| GLC 2000 | 68.6 | Confidence values statistical sampling | SPOT4 VEGETATION | 非监督分类 | 1 km | 1999-2000 | FAO LCCS (23 classes) | http://bioval.jrc.ec.europa.eu/products/glc2000/products.php | Bartholomé等[26] |
| IGBPDIS | 66.9 | Statistical sampling of validation working group | AVHRR | 非监督分类 | 1 km | 1992-1993 | USGS IGBP (17 classes) | http://edc2.usgs.gov/glcc/tabgoode_globe.php | Loveland等[27] |
| UMD | 65.0 | Evaluated using other digital datasets | AVHRR | 非监督分类、 决策树分类 | 1 km | 1992-1993 | Simplified IGBP (14 classes) | http://www.landcover.org/data/landcover/index.shtml | Hansen等[28] |
| MCD12Q1 | 74.8 | Cross-validation | MODIS | 监督分类、决策树分类、神经网络 | 500 m | 2013 | IGBP (17 classes) | http://e4ftl01.cr.usgs.gov/MOTA/MCD12Q1.051/ | Friedl等[29,30] |
| GlobCover | 67.5 | Statistical sampling expert's judgement | MERIS FR | 监督分类、 非监督分类 | 300 m | 2009 | UN LCCS (22 classes) | http://due.esrin.esa.int/globcover/ | Bontemps等[31] |
| CCI-LC | 74.1 | Sampling-based labeling approach | MERIS Full and Reduced Resolution/ SPOT | 非监督分类 | 300 m | 2008-2012 | UN LCCS (22 classes) | http://maps.elie.ucl.ac.be/CCI/viewer/index.php | Belgium等[32] |
| GlobeLand 30 | 80.0 | Knowledge-based interactive verification | Landsat TM, ETM7, HJ-1A/b/ | 基于像元、对象和知识规则分类 | 30 m | 2010 | 11 classes | http://www.globallandcover.com | Chen等[33] |
新窗口打开
2.2 样本点数据
本文所用样本数据来自野外实地调查、基于空间采样的湖泊样点调查和冰川样点调查,共计6851个样本点（表2）。其中在2012-2014年羌塘高原土地覆被类型野外调查中,获取2991个样本数据,样本点主要分布在陆面交通条件较好的区域;基于谷歌地球影像和相片标记的湖泊样本点1157个,范围大小为2 km×2 km;冰川样本点数据源自寒区旱区科学数据中心中国第二次冰川编目数据集[25],计2703个,范围为2 km×2 km,样点的空间分布如图1所示。Tab. 2
表2
表2羌塘高原样本点概况
Tab. 2Class description of the filed sample in Qiangtang Plateau
| 类型 | 样本点数量 | 类型定义 | 类型 | 样本点数量 | 类型定义 |
|---|---|---|---|---|---|
| 高寒草甸 | 777 | 由寒冷中生多年生草本植物为主的植物群落覆盖区域,本研究区主要指藏北嵩草草甸、小嵩草（高山嵩草）草甸覆盖区 | 湖泊 | 1157 | 指自然条件下形成的积水区常年水位以下的土地 |
| 高寒草原 | 1245 | 具有一定御寒能力的、旱生的多年生草本植物和小半灌木植物占优势的植物群落覆盖区域 | 沼泽湿地 | 80 | 指覆盖着水（淡水、半咸水或咸水）与草本或木本植物的广阔区域,是介于陆地和水体之间的过渡带 |
| 稀疏植被 | 29 | 分布在连续植物覆盖的植被以上至永久雪线之间的、由适应严寒生境的寒旱生或寒冷中旱生多年生轴根性杂类草或以垫状植物或地衣苔藓等构成的盖度在5%~40%的植被区域。如蚤缀、点地梅垫状植被分布区域 | 居民建设用地 | 35 | 指被建筑物覆盖的土地类型 |
| 半灌木或矮半灌木荒漠 | 662 | 半灌木、矮半灌木（驼绒藜、木亚菊、蒿）荒漠、垫状驼绒藜荒漠广泛分布区域 | 裸地 | 125 | 指裸地、沙地、岩石、盐碱地,植被覆盖度不超过10 % |
| 河流 | 38 | 指自然形成的沿着地表长条状槽形洼地 | 雪被与冰川 | 2703 | 指常年由积雪或者冰覆盖的土地类型 |
| 合计 | 6851 |
新窗口打开
显示原图|下载原图ZIP|生成PPT图1羌塘高原样本点空间分布
-->Fig.1Spatial distribution of the filed investigated sample points in Qiangtang Plateau
-->
2.3 数据处理与方法
2.3.1 土地覆被数据预处理 七套土地覆被数据的投影统一为“Albers_Conic_Equal_Area”。数据的空间分辨率保持不变,具体原因为：通过重采样的方法,如面积占优法、最近邻法等[12],其结果会降低数据本身的质量,且数据产品分辨率范围跨度大（30 m~1 km）,难以选择折中且对数据影响小的统一分辨率;其次,由于羌塘高原特殊的地理环境和植被特征,从草地、荒漠、裸地主要土地覆被类型中筛选出的代表性样本点,能保持周围类别单一,成片分布范围大于最大像元的大小（1 km）,基本不会影响像元大小造成的数据评价结果精度。因此,在不影响样本点对数据评价结果的条件下,为了尽可能减少对数据本身精度的影响,本文决定只对其进行投影转换处理,不对空间分辨率进行处理。2.3.2 分类系统对应关系 各类土地覆被数据中土地覆被类型的划分及相关标准的制定存在一定差异,这使得不同数据间的比较难以进行[4]。Giri等[34]于2005年、Ran等[35]于2010年和Herold等[4]于2008年的相关研究中通过建立分类系统的衔接关系,实现了不同覆被类型的对应。参照已有的分类系统衔接工作[12,36,37],并结合实地考察和各数据产品类型的具体定义,编制了基于类型样点调查的羌塘高原土地覆被分类系统,包括8个一级类型10个二级类型。八套数据的四套分类系统与该分类系统具体的对应关系如表3所示。
Tab. 3
表3
表3羌塘高原地区不同土地覆被数据与样本点分类系统间的类别对应关系
Tab. 3Corresponding relationships of classes in different classification system between eight land cover data sets and ground data
| 类型 | IGBP (IGBPDIS、MCD12Q1、UMD) | FAO LCCS (GLC 2000、GlobCover、CCI-LC) | GlobeLand 30 | ||
|---|---|---|---|---|---|
| 序号 | 一级 | 二级 | |||
| 1 | 草地 | 草地 | 草原、草甸 | 草地 | |
| 2 | 高寒草甸 | - | 草甸 | - | |
| 3 | 高寒草原 | 草地 | 草原 | 草地 | |
| 4 | 稀疏植被 | 稀疏植被 | - | 自然植被与农田镶嵌类型、稀疏植被 | - |
| 5 | 半灌木或矮半灌木荒漠 | 半灌木或矮半灌木荒漠 | 稀疏灌木林地 | 荒漠草地、灌丛、落叶灌丛 | 灌木林地 |
| 6 | 水体 | 水体 | 水体 | 水体 | |
| 7 | 河流 | - | - | - | |
| 8 | 湖泊 | - | - | - | |
| 9 | 沼泽湿地 | 沼泽湿地 | 湿地 | 稀疏草本或木本湿地 | 湿地 |
| 10 | 居民建设用地 | 居民建设用地 | 建设用地 | 建设用地 | 人造地表 |
| 11 | 裸地 | 裸地 | 裸地 | 裸地、砾石裸地、松散裸地 | 裸地 |
| 12 | 雪被和冰川 | 雪被和冰川 | 雪被与冰川 | 永久雪被与冰川 | 永久雪被与冰川 |
| 13 | 无 | - | 常绿针叶林、常绿阔叶林、落叶针叶林、落叶阔叶林、混交林、郁闭灌木林地、森林稀疏草原、稀树草原、农田、农田与自然植被镶嵌类型 | 旱地、灌溉或季节性水淹农田、农田与自然植被镶嵌类型、常绿针叶林 | 农田、乔木林地 |
新窗口打开
2.3.3 验证方法 本文采用混淆矩阵的方法,以总体精度、制图精度和用户精度为指标衡量土地覆被数据产品的质量,计算公式分别为[38, 39]：
式中：OA为总体精度,是所有被正确识别的像元数T与总样本数X的比值;PAi为某地类i的生产者精度,或者制图精度;Ti为被正确识别为第i类地物的像元数;Xi为实际为第i类地物的样本数;Vi为被分为第i类地物的样本数;UAi为某地类i的用户精度。
3 结果分析
3.1 一致性特征
3.1.1 空间一致性 根据野外调查和相关资料记载[40,41],草地、荒漠和裸地是羌塘高原地区主要的一级土地覆被类型,又以草地类型占比最大。草地主要分布在中部和南部,荒漠主要分布在北部,裸地主要分布在西北部的草地与荒漠过渡带。羌塘高原主要的3种土地覆被一级类型在不同的数据产品中差异明显（图2）。CCI-LC、GlobeLand 30和GLC 2000数据主要覆被类型是草地,面积比例分别为70.10%、67.69%、62.92%、64.42%,能较好地描述羌塘高原中部和东南部的草地空间分布。MCD12Q1数据中裸地和草地面积比例接近,分别为40.91%和44.96%,但其他三套数据的空间分布与调查结果有明显偏差。
显示原图|下载原图ZIP|生成PPT图2七套数据产品在羌塘高原地区一级类型的空间分布特征 注：图中及下文各图中的不一致类别均指研究区现实中不存在的土地覆被类型,而各对应的数据产品在本区域中出现的土地覆被类型。
-->Fig. 2Distribution of eight data sets at Level I of the classification on the Qiangtang Plateau
-->
从七套数据草地二级类型的空间分布中可以发现（图3）,高寒草甸比例非常少,高寒草原空间分布特征和数据精度与一级类型中草地分布一致,CCI-LC、GlobeLand 30、GLC 2000数据相对能较好得识别草地类型。而各数据湖泊类型的空间分布特征基本类似,识别度较高。但与Zhang等[42]2015年目视解译的羌塘高原地区面积大于10 km2的湖泊数据对比,七套土地覆被数据的湖泊面积明显偏小。本文土地覆被数据（1992-2010年）湖泊的面积范围在14712.00~28039.18 km2之间,Zhang等（1991-2013年）获取的湖泊数据面积范围为24502.47~30737.67 km2。
显示原图|下载原图ZIP|生成PPT图3七套土地覆被数据产品在羌塘高原地区二级类型的空间分布特征
-->Fig. 3Distribution of seven data sets at Level II of the classification on the Qiangtang Plateau
-->
3.1.2 点位一致性 综合数据一致性和样本点验证的结果来看,一致性程度低,则实际土地覆被类型准确率低;一致性程度高,与实际土地覆被类型之间的关系并不明显（图4a）。总体上,在6851个样本点中,七套数据一致性程度低,主要集中在3和4之间,占54.9%（图4b）。数据间一致性低、数据与样本点间准确度也低（LL）的像元比例最大,占63.86%。即在大部分验证点上,各数据产品对类型判断不一致且判断准确的比例低。数据间一致性在空间上,未呈现明显区域差异特征。具体类型中（图4c）,一致性和准确度高（HH）的类型主要集中在高寒草地和雪被冰川类型上。
显示原图|下载原图ZIP|生成PPT图4土地覆被数据一致性与准确性分布图注：图b中,纵坐标值为1表示在该样本点上7套数据的分类结果各不相同,值为2表示在该样本点上有2套产品的分类结果是一致的,……,值为7表示有7套产品在该样本点上的分类结果全部一致。
-->Fig. 4Agreement and accuracy of land cover data sets
-->
3.2 总体精度
七套数据一级土地覆被类型在羌塘高原地区总体精度均低于56%,可划分为4个等级,由高到低分别为：50%以上,40%~50%,20%~40%以及20%以下。从表4可以看出各数据总体精度情况具体情况,第Ⅰ级：GlobeLand 30（55.09%）和CCI-LC（53.92%）;第Ⅱ级：GLC 2000（49.97%）和CASLU（40.48%）;第Ⅲ级：GlobCover（31.88%）和MCD12Q1（24.61%）;第Ⅳ级：UMD（5.98%）和IGBPDIS（11.76%）。七套数据产品二级类型总体精度为6%~47%,精度等级与一级类型一致。Table 4
表4
表4七套数据产品中不同类型面积比例及总体精度（%）
Table 4Area proportion and overall accuracy of different land cover types of the seven data products (%)
| 类型 | CCI-LC | GLC 2000 | GlobCover | GlobeLand30 | IGBPDIS | MCD12Q1 | UMD | |
|---|---|---|---|---|---|---|---|---|
| 本文 研究区 | 稀疏植被 | 2.00 | 0.00 | 2.51 | 0.00 | 0.00 | 0.00 | 0.00 |
| 湿地 | 0.07 | 0.00 | 0.00 | 0.31 | 0.31 | 0.06 | 0.00 | |
| 建设用地 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | |
| 雪被与冰川 | 2.25 | 2.49 | 1.64 | 1.19 | 1.19 | 1.25 | 0.00 | |
| 草地 | 70.51 | 67.69 | 68.84 | 86.12 | 40.47 | 3.53 | 19.75 | |
| 高寒荒漠 | 1.10 | 16.62 | 0.00 | 2.34 | 2.34 | 9.49 | 68.50 | |
| 水体 | 4.00 | 2.77 | 4.07 | 4.40 | 4.40 | 3.32 | 4.09 | |
| 裸地 | 20.06 | 9.28 | 51.31 | 24.07 | 24.07 | 40.91 | 23.88 | |
| 一级类型 总体精度 | 53.92 | 49.97 | 31.88 | 55.09 | 11.76 | 24.61 | 5.98 | |
| 二级类型 总体精度 | 46.23 | 39.41 | 26.23 | 46.55 | 10.61 | 21.46 | 6.00 | |
| 等级 | Ⅰ | Ⅱ | Ⅲ | Ⅰ | Ⅳ | Ⅲ | Ⅳ | |
| 国际报道 | 总体精度 | 74.40 | 68.60 | 67.50 | 80.03 | 66.90 | 74.80 | 65.00 |
| 参考文献 | 文献[32] | 文献[26] | 文献[31] | 文献[33] | 文献[27] | 文献[29, 30] | 文献[28] |
新窗口打开
3.3 不同类型精度
3.3.1 生产者精度 七套数据产品的生产者精度普遍偏低。草地类型中,GLC 2000数据精度为46.19%,CCI-LC数据精度为43.57%,其他数据的制图精度均小于40%。在荒漠类型中,IGBPDIS数据制图精度为84.44%,但其总体精度仅为11.76%。裸地类型识别中,七套数据制图精度均小于50%,最高的仅为39.20%（MCD12Q1）。CCI-LC数据在雪被与冰川类型识别上制图精度为80.01%,且错分概率为7.2%,优于其他数据产品。而各数据在稀疏植被、湿地、水体类型中的制图精度均低于40%（图5a）。
显示原图|下载原图ZIP|生成PPT图5一级土地覆被类型的生产者精度和用户精度
-->Fig. 5Producer's accuracy and user's accuracy of seven land cover data sets at Level I of the classification
-->
3.3.2 用户精度 三大主要类型用户精度均低于50%,精度相对较高的类型主要是水体和冰川。在水体类型上,GlobeLand 30准确度为92.10%,漏分概率为64.74%。GlobCover准确度虽然为93.33%,漏分概率为93.95%。可见,GlobeLand 30在水体识别上整体精度更高。在冰川雪被类型上,CCI-LC数据精度更高。GlobeLand 30虽然识别准确率达到97.35%,但其漏分概率（27.89%）,高于CCI-LC数据（图5b）。
3.3.3 面积估算精度 面积估算精度表示生产者精度与用户精度差值,在生产者精度和用户精度均较高的条件下,其值越接近表明该类型面积估算精度越高。经计算,制图精度和用户精度均较高的有GLC 2000、CCI-LC和GlobeLand 30四套数据的草地类型（图5）。其中,面积精度最高的为GlobeLand 30数据。其他类型中,无同时满足制图精度、用户精度和面积精度高的数据（表5）。
Tab. 5
表5
表5七套数据产品中不同类型面积估算精度（%）
Tab. 5Accuracy of area estimation of different land cover types of the seven data products (%)
| 数据 | 稀疏植被 | 湿地 | 建设用地 | 雪被与冰川 | 草地 | 高寒荒漠 | 水体 | 裸地 |
|---|---|---|---|---|---|---|---|---|
| CCI-LC | 0.00 | 0.00 | 0.00 | 12.70 | 12.32 | 7.79 | 73.25 | 14.48 |
| GlobalLand 30 | 0.00 | 8.75 | 26.83 | 25.24 | 4.41 | 0.00 | 57.96 | 34.43 |
| GLC 2000 | 0.00 | 0.00 | 0.00 | 36.46 | 15.00 | 19.17 | 40.85 | 0.68 |
| GlobCover | 0.00 | 0.00 | 0.00 | 42.87 | 1.74 | 0.00 | 87.48 | 21.40 |
| MCD12Q1 | 0.00 | 64.17 | 0.00 | 55.69 | 1.18 | 3.55 | 81.24 | 37.59 |
| IGBPDIS | 0.00 | 0.00 | 0.00 | 0.00 | 21.53 | 73.82 | 30.03 | 0.70 |
| UMD | 0.00 | 0.00 | 0.00 | 0.00 | 4.88 | 42.86 | 42.54 | 9.90 |
新窗口打开
3.4 不同类型混淆
利用混淆矩阵,获得数据和样本点在相同位置土地覆被类型对应关系,选取六套总体精度高于20%的数据产品进行统计分析,地类混淆情况如图6所示,各类型与草地和裸地类型混淆严重。草地类型中,CCI-LC数据9.20%混分为裸地,少部分混分为其他类型。GlobeLand 30中,37.83%混分为裸地。GLC 2000数据18.23%混分为荒漠。裸地类型中,GlobeLand 30中24.59%混分为草地。MCD12Q1中,13.11%混分为荒漠。雪被与冰川类型中,CCI-LC中混分为草地和裸地比例分别为10.69%和8.77%。GlobeLand 30雪被与冰川主要混分为裸地（20.35%）,另有少部分混分为其他类型。GLC 2000数据的雪被与冰川有44.98%混分为其他类型。此外,荒漠、稀疏植被、建设用地、湿地、水体主要混分为草地、裸地类型。
显示原图|下载原图ZIP|生成PPT图6土地覆被一级类型混淆比例
-->Fig. 6Confusion of Level I of the classification between the six land cover datasets
-->
4 讨论
本文评估的七套数据采用了三套分类系统,即FAO LCCS、IGBP和GlobeLand30。在分析系统对应的过程中,主要问题表现为以下几方面：① UMD分类系统虽然采用了IGBP的分类方案,但UMD土地覆盖数据的分类系统是为SIB模型设计的,SIB的分类方案没有湿地和冰雪类,而是将永久性冰雪类包括在裸地里,而没有湿地类型,这可能会降低UMD数据的精度评价结果。② 羌塘高原的主题范围内基本无林地和耕地类型[23],但七套数据产品中均有林地和耕地类型,尤其IGBPDIS 和UMD数据在羌塘高原的核心区出现了大片林地和耕地类型（图2）,偏离了真实情况。以上问题说明,不仅数据集分类系统的转换过程会影响数据精度[43],分类原则和类型内涵的不同,也是造成本文区的遥感分类解译结果差异因素之一。而出现偏离真实类别的分类系统的原因很大程度是由于缺乏野外实测资料。因此,这些数据的分类系统需要进一步考察修正,以减少分类系统制定过程中的模糊概念。羌塘高原植被类型和高原环境特征对土地覆被数据分类有重要影响。羌塘高原特殊的植被类型,如草地、裸地、荒漠类型波谱曲线、影像色调、纹理等标识差异小,易造成类型混淆。其次,羌塘高原典型的高原环境对土地覆被提取具有重要的影响。羌塘高原的植被生长季远短于其他低海拔地区[44,45],主要集中在夏季,其他月份难以监测植被类型信息,但土地覆被数据使用的各幅影像的月份可能不一致,如使用非生长季期间的遥感影像数据,则会降低植被类型解译结果。
本文主要不确定性表现为：样本点采集与遥感数据年份存在时间差异。本文样本点数据来自2013-2016年野外调查,选用的七套数据时间跨度约为20年（1992年/1993年-2013年）。但据实地考察发现,羌塘高原大面积属于无人区,主要受自然因素影响,该地区整体上土地覆被类型变化幅度相对较小。其次,样本点空间分布不够均匀。由于环境条件极其恶劣,本文野外获取的样本点数据主要分布在陆面交通条件较好的区域,空间分布不够均匀对精度评价结果会有一定影响,但作为该地区首批基础调查数据之一,是评估该地区土地覆被数据精度的重要参考。
5 结论
在羌塘高原地区,七套数据产品（GLC 2000、IGBPDIS、UMD、MCD12Q1、GlobCover、CCI-LC和GlobeLand 30）中,精度最高的一级类型仅为55.09%左右,未达到可广泛应用的精度要求[1,2]。从具体类型来看,雪被与冰川数据类型制图精度和用户精度整体上高于其他数据,其中最高的是CCI-LC数据,用户精度为92.80%,制图精度为80.10%。GlobeLand 30数据的草地类型面积估算精度最高。本文评估的七套大尺度土地覆被数据产品在高海拔的羌塘高原寒旱地区精度低,无论从现状分析,还是作为区域土地覆被变化及其生态和环境影响研究的基础数据而言,高原地区土地覆被数据质量都亟待提高。建议如下：① 加强野外实地土地覆被类型的光谱测定、植被样地调查、土壤与土地利用调查及环境要素调查,确定类型解译标志和标准,及梳理类型内涵与定义,调整并完善分类系统,修订和完善遥感解译流程和方法以提升解译能力;② 采用最新的高分辨率影像、包含物候信息的时序数据（SPOT VGT、HJ CCD以及MODIS）,尤其是结合2000年以来MODIS的年内与年际间多期连续的时序数据,并结合野外实际观测分析结果,针对某土地覆被类型在进行深化和创新性研究。
致谢：对野外土地利用覆被调查中给予指导和帮助的中国科学院植物研究所郭柯研究员表示感谢,对参与野外数据采集的赵志龙、谷昌军、郑海朋、王宇坤等表示感谢,对数据分析和论文撰写中给予帮助和提出宝贵建议的王兆锋、聂勇等老师表示感谢。
The authors have declared that no competing interests exist.
参考文献 原文顺序
文献年度倒序
文中引用次数倒序
被引期刊影响因子
| [1] | Abstract Abstract: We identified 100 scientific questions that, if answered, would have the greatest impact on conservation practice and policy. Representatives from 21 international organizations, regional sections and working groups of the Society for Conservation Biology, and 12 academics, from all continents except Antarctica, compiled 2291 questions of relevance to conservation of biological diversity worldwide. The questions were gathered from 761 individuals through workshops, email requests, and discussions. Voting by email to short-list questions, followed by a 2-day workshop, was used to derive the final list of 100 questions. Most of the final questions were derived through a process of modification and combination as the workshop progressed. The questions are divided into 12 sections: ecosystem functions and services, climate change, technological change, protected areas, ecosystem management and restoration, terrestrial ecosystems, marine ecosystems, freshwater ecosystems, species management, organizational systems and processes, societal context and change, and impacts of conservation interventions. We anticipate that these questions will help identify new directions for researchers and assist funders in directing funds. |
| [2] | Land-cover (LC) products, especially at the regional and global scales, comprise essential data for a wide range of environmental studies affecting biodiversity, climate, and human health. This review builds on previous compartmentalized efforts by summarizing 23 global and 41 regional LC products. Characteristics related to spatial resolution, overall accuracy, time of data acquisition, sensor used, classification scheme and method, support for LC change detection, download location, and key corresponding references are provided. Operational limitations and uncertainties are discussed, mostly as a result of different original modelling outcomes. Upcoming products are presented and future prospects towards increasing usability of different LC products are offered. Despite the common realization of product usage by non-experts, the remote-sensing community has not fully addressed the challenge. Algorithmic development for the effective representation of inherent product limitations to facilitate proper usage by non-experts is necessary. Further emphasis should be placed on international coordination and harmonization initiatives for compatible LC product generation. We expect the applicability of current and future LC products to increase, especially as our environmental understanding increases through multi-temporal studies. |
| [3] | We have produced the first 30 m resolution global land-cover maps using Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper Plus (ETM+) data. We have classified over 6600 scenes of Landsat TM data after 2006, and over 2300 scenes of Landsat TM and ETM+ data before 2006, all selected from the green season. These images cover most of the world's land surface except Antarctica and Greenland. Most of these images came from the United States Geological Survey in level L1T (orthorectified). Four classifiers that were freely available were employed, including the conventional maximum likelihood classifier (MLC), J4.8 decision tree classifier, Random Forest (RF) classifier and support vector machine (SVM) classifier. A total of 91,433 training samples were collected by traversing each scene and finding the most representative and homogeneous samples. A total of 38,664 test samples were collected at preset, fixed locations based on a globally systematic unaligned sampling strategy. Two software tools, Global Analyst and Global Mapper developed by extending the functionality of Google Earth, were used in developing the training and test sample databases by referencing the Moderate Resolution Imaging Spectroradiometer enhanced vegetation index (MODIS EVI) time series for 2010 and high resolution images from Google Earth. A unique land-cover classification system was developed that can be crosswalked to the existing United Nations Food and Agriculture Organization (FAO) land-cover classification system as well as the International Geosphere-Biosphere Programme (IGBP) system. Using the four classification algorithms, we obtained the initial set of global land-cover maps. The SVM produced the highest overall classification accuracy (OCA) of 64.9% assessed with our test samples, with RF (59.8%), J4.8 (57.9%), and MLC (53.9%) ranked from the second to the fourth. We also estimated the OCAs using a subset of our test samples (8629) each of which represented a homogeneous area greater than 500 mx500 m. Using this subset, we found the OCA for the SVM to be 71.5%. As a consistent source for estimating the coverage of global land-cover types in the world, estimation from the test samples shows that only 6.90% of the world is planted for agricultural production. The total area of cropland is 11.51% if unplanted croplands are included. The forests, grasslands, and shrublands cover 28.35%, 13.37%, and 11.49% of the world, respectively. The impervious surface covers only 0.66% of the world. Inland waterbodies, barren lands, and snow and ice cover 3.56%, 16.51%, and 12.81% of the world, respectively. |
| [4] | Many investigators need and use global land cover maps for a wide variety of purposes. Ironically, after many years of very limited availability, there are now multiple global land cover maps and it is not readily apparent (1) which is most useful for particular applications or (2) how to combine the different maps to provide an improved dataset. The existing global land cover maps at 1km spatial resolution have arisen from different initiatives and are based on different remote sensing data and employed different methodologies. Perhaps more significantly, they have different legends. As a result, comparison of the different land cover maps is difficult and information about their relative utility is limited. In an attempt to compare the datasets and assess their strengths and weaknesses we harmonized the thematic legends of four available coarse-resolution global land cover maps (IGBP DISCover, UMD, MODIS 1-km, and GLC2000) using the LCCS-based land cover legend translation protocols. Analysis of the agreement among the global land cover maps and existing validation information highlights general patterns of agreement, inconsistencies and uncertainties. The thematic classes of Evergreen broadleaf trees, Snow and Ice, and Barren show high producer and user accuracy and good agreement among the datasets, while classes of mixed tree types show high commission errors. Overall, the results show a limited ability of the four global products to discriminate mixed classes characterized by a mosaic of trees, shrubs, and herbaceous vegetation. There is a strong relationship between class accuracy, spatial agreement among the datasets, and the heterogeneity of landscapes. Suggestions for future mapping projects include careful definition of mixed unit classes, and improvement in mapping heterogeneous landscapes. |
| [5] | Given the advances in remotely sensed imagery and associated technologies, several global land cover maps have been produced in recent times including IGBP DISCover, UMD Land Cover, Global Land Cover 2000 and GlobCover 2009. However, the utility of these maps for specific applications has often been hampered due to considerable amounts of uncertainties and inconsistencies. A thorough review of these global land cover projects including evaluating the sources of error and uncertainty is prudent and enlightening. Therefore, this paper describes our work in which we compared, summarized and conducted an uncertainty analysis of the four global land cover mapping projects using an error budget approach. The results showed that the classification scheme and the validation methodology had the highest error contribution and implementation priority. A comparison of the classification schemes showed that there are many inconsistencies between the definitions of the map classes. This is especially true for the mixed type classes for which thresholds vary for the attributes/discriminators used in the classification process. Examination of these four global mapping projects provided quite a few important lessons for the future global mapping projects including the need for clear and uniform definitions of the classification scheme and an efficient, practical, and valid design of the accuracy assessment. |
| [6] | Global land cover (GLC) maps and assessments of their accuracy provide important information for different user communities. To date, there are several GLC reference datasets which are used for assessing the accuracy of specific maps. Despite significant efforts put into generating them, their availability and role in applications outside their intended use have been very limited. This study analyses metadata information from 12 existing and forthcoming GLC reference datasets and assesses their characteristics and potential uses in the context of 4 GLC user groups, i.e., climate modellers requiring data on Essential Climate Variables (ECV), global forest change analysts, the GEO Community of Practice for Global Agricultural Monitoring and GLC map producers. We assessed user requirements with respect to the sampling scheme, thematic coverage, spatial and temporal detail and quality control of the GLC reference datasets. Suitability of the datasets is highly dependent upon specific applications by the user communities considered. The LC-CCI, GOFC-GOLD, FAO-FRA and Geo-Wiki datasets had the broadest applicability for multiple uses. The re-usability of the GLC reference datasets would be greatly enhanced by making them publicly available in an expert framework that guides users on how to use them for specific applications. |
| [7] | The production of thematic maps, such as those depicting land cover, using an image classification is one of the most common applications of remote sensing. Considerable research has been directed at the various components of the mapping process, including the assessment of accuracy. This paper briefly reviews the background and methods of classification accuracy assessment that are commonly used and recommended in the research literature. It is, however, evident that the research community does not universally adopt the approaches that are often recommended to it, perhaps a reflection of the problems associated with accuracy assessment, and typically fails to achieve the accuracy targets commonly specified. The community often tends to use, unquestioningly, techniques based on the confusion matrix for which the correct application and interpretation requires the satisfaction of often untenable assumptions (e.g., perfect coregistration of data sets) and the provision of rarely conveyed information (e.g., sampling design for ground data acquisition). Eight broad problem areas that currently limit the ability to appropriately assess, document, and use the accuracy of thematic maps derived from remote sensing are explored. The implications of these problems are that it is unlikely that a single standardized method of accuracy assessment and reporting can be identified, but some possible directions for future research that may facilitate accuracy assessment are highlighted. |
| [8] | Freely available satellite imagery will improve science and environmental-monitoring products, say Michael A. Wulder and Nicholas C. Coops. |
| [9] | The MODIS Land team is producing a suite of global land products whose uncertainty will be estimated through validation activities. The MODIS Land team will base its validation work on the comparison of its products to similar products derived from independent sources. The independent products will be derived from a combination of in situ data and imagery from airborne and spaceborne sensors. Since in situ and image data can often serve to validate more than one product and sensor, the MODIS Land Discipline Team's validation strategy has focused on data collection and analysis at the EOS Land Validation Core Sites. Initial characterization of these sites is presented, as well as an overview of the on-line access to imagery and field data collected over these sites. The data and resources available through this work are available to the science community for continued validation and scientific investigations. This paper describes the results of a 4-year effort to develop the infrastructure to allow timely and comprehensive validation of EOS land products. |
| [10] | Global change issues are high on the current international political agenda. A variety of global protocols and conventions have been established aimed at mitigating global environmental risks. A system for monitoring, evaluation and compliance of these international agreements is needed, with each component requiring comprehensive analytical work based on consistent datasets. Consequently, scientists and policymakers have put faith in earth observation data for improved global analysis. Land cover provides in many aspects the foundation for environmental monitoring [FAO, 2002a. Proceedings of the FAO/UNEP Expert Consultation on Strategies for Global Land Cover Mapping and Monitoring. FAO, Rome, Italy, 38 pp.]. Despite the significance of land cover as an environmental variable, our knowledge of land cover and its dynamics is poor [Foody, G.M., 2002. Status of land cover classification accuracy assessment. Rem. Sens. Environ. 80, 185 201]. This study compares four satellite derived 1 km land cover datasets freely available from the internet and in wide use among the scientific community. Our analysis shows that while these datasets have in many cases reasonable agreement at a global level in terms of total area and general spatial pattern, there is limited agreement on the spatial distribution of the individual land classes. If global datasets are used at a continental or regional level, agreement in many cases decreases significantly. Reasons for these differences are many anging from the classes and thresholds applied, time of data collection, sensor type, classification techniques, use of in situ data, etc., and make comparison difficult. Results of studies based on global land cover datasets are likely influenced by the dataset chosen. Scientists and policymakers should be made aware of the inherent limitations in using current global land cover datasets, and would be wise to utilise multiple datasets for comparison. |
| [11] | |
| [12] | Land cover dynamics at the African continental scale is of great importance for global change studies. Actually, four satellite-derived land cover maps of Africa now available, e.g. ECOCLIMAP, GLC2000, MODIS and GLOBCOVER, are based on images acquired in the 2000s. This study aims at stressing the compliances and the discrepancies between these four land cover classifications systems. Each of them used different mapping initiatives and relies on different mapping standards, which supports the present investigation. In order to do a relative comparison of the four maps, a preamble was to reconcile their thematic legends into more aggregated categories after a projection into the same spatial resolution. Results show that the agreement between the four land cover products is between 56 and 69%. While all these land cover datasets show a reasonable agreement in terms of surface types and spatial distribution patterns, mapping of heterogeneous landscapes in the four products is not very successful. Land cover products based on remote sensing imagery can indeed significantly be improved by using smarter algorithms, better timing of image acquisition, improved class definitions. Either will help to improve the accuracy of future land cover maps at the African continental scale. Data producers may use the areas of spatial agreement for training area selection while users might need to verify the information in the areas of disagreement using additional data sources. |
| [13] | Many investigators use global land cover (GLC) maps for different purposes, such as an input for global climate models. The currentGLC maps used for such purposes are based on different remote sensing data, methodologies and legends. Consequently,comparison of GLC maps is difficult and information about their relative utility is limited. The objective of this study is to analyseand compare the thematic accuracies of GLC maps (i.e., IGBP-DISCover, UMD, MODIS, GLC2000 and SYNMAP) at 1 kmresolutions by (a) re-analysing the GLC2000 reference dataset, (b) applying a generalized GLC legend and (c) comparing theirthematic accuracies at different homogeneity levels. The accuracy assessment was based on the GLC2000 reference dataset with1253 samples that were visually interpreted. The legends of the GLC maps and the reference datasets were harmonized into 11general land cover classes. There results show that the map accuracy estimates vary up to 10-16% depending on the homogeneity ofthe reference point (HRP) for all the GLC maps. An increase of the HRP resulted in higher overall accuracies but reduced accuracyconfidence for the GLC maps due to less number of accountable samples. The overall accuracy of the SYNMAP was the highest atany HRP level followed by the GLC2000. The overall accuracies of the maps also varied by up to 10% depending on the definitionof agreement between the reference and map categories in heterogeneous landscape. A careful consideration of heterogeneouslandscape is therefore recommended for future accuracy assessments of land cover maps.* Corresponding author1. INTRODUCTIONThe consistent and continuous observation of land cover is oneof the most important foundations for understanding the Earth environment and ecosystems (Verburg et al., 2011). Currently,several global land cover datasets (GLC) have been developedand these datasets are evolving towards higher spatial resolution(Gong et al., 2013; Mora et al., 2014) . Most GLC maps weredeveloped by individual groups as one-time efforts and thesubsequent mapping standards reflect the varied interests,requirements and methodologies of the originating programs(Herold et al., 2006). These differences of GLC maps and theeffects of their quality on the model outcome are not alwaysconsidered when selecting a map as an input for specificmodeling applications (Verburg et al., 2011). Uncertainties ofGLC maps can result in considerable differences in modelingoutcomes (Hibbard et al., 2010; Nakaegawa, 2011; Verburg etal., 2011).The accuracies of GLC maps are assessed using independentvalidation datasets and regional maps or cross validated againsttraining datasets. The results of accuracy assessments ofprevious maps indicate that overall area-weighted accuracy isaround 70% for the existing GLC maps (Defourny et al., 2012).However, the use of different approaches in the GLC mapproduction (e.g., classification scheme, data sources andalgorithms) as well as in validation data collection (e.g.,sampling scheme, data source and method of referenceclassification) raise inconsistency issues and make mapcomparisons difficult. Several comparative analyses of landcover maps were conducted at regional levels |
| [14] | This paper presents a rigorous validation of five widely used global land cover products, i.e., GLCC (Global Land Cover Characterization), UMd (University of Maryland land cover product), GLC2000 (Global Land Cover 2000 project data), MODIS LC (Moderate Resolution Imaging Spectro-radiometer Land Cover product) and GlobCover (GLOBCOVER land cover product), and a national land cover map GLCD-2005 (Geodata Land Cover Dataset for year 2005) against an independent reference data set over China. The land cover reference data sets in three epochs (1990, 2000, and 2005) were collected on a web-based prototype system using a sampling-based labeling approach. Results show that, in China, the highest overall accuracy is observed in GLCD-2005 (72.3%), followed by MODIS LC (68.9%), GLC2000 (65.2%), GlobCover (57.7%) and GLCC (57.2%), while UMd has the lowest accuracy (48.6%); all of the products performed best in representing "Trees" and "Others", well with "Grassland" and "Cropland", but problematic with "Water" and "Urban" across China in general. Moreover, in respect of GLCD-2005, there are significant accuracy differences across seven geographical locations of China, ranging from 46.3% in the Southwest, 77.5% in the South, 79.2% in the Northwest, 80.8% in the North, 81.8% in the Northeast, 82.6% in the Central, to 89.0% in the East. This study indicates that a regionally focused land cover map would in fact be more accurate than extracting the same region from a globally produced map. |
| [15] | Land cover mapping in mountainous areas is a notoriously challenging task due to the rugged terrain and high spatial heterogeneity of land surfaces as well as the frequent cloud contamination of satellite imagery. Taking Southwestern China (a typical mountainous region) as an example, this paper established a new HC-MMK approach (Hierarchical Classification based on Multi-source and Multi-temporal data and geo-Knowledge), which was especially designed for land cover mapping in mountainous areas. This approach was taken in order to generate a 30 m-resolution land cover product in Southwestern China in 2010 (hereinafter referred to as CLC-SW2010). The multi-temporal native HJ (HuanJing, small satellite constellation for disaster and environmental monitoring) CCD (Charge-Coupled Device) images, Landsat TM (Thematic Mapper) images and topographical data (including elevation, aspect, slope, etc.) were taken as the main input data sources. Hierarchical classification tree construction and a five-step knowledge-based interactive quality control were the major components of this proposed approach. The CLC-SW2010 product contained six primary categories and 38 secondary categories, which covered about 2.33 million km(2) (accounting for about a quarter of the land area of China). The accuracies of primary and secondary categories for CLC-SW2010 reached 95.09% and 87.14%, respectively, which were assessed independently by a third-party group. This product has so far been used to estimate the terrestrial carbon stocks and assess the quality of the ecological environments. The proposed HC-MMK approach could be used not only in mountainous areas, but also for plains, hills and other regions. Meanwhile, this study could also be used as a reference for other land cover mapping projects over large areas or even the entire globe. |
| [16] | |
| [17] | The accuracy of a land cover classification is the degree to which the map land cover agrees with the reference land cover classification (i.e. ground condition). The basic sampling designs historically implemented for map accuracy assessment have served well for the error matrix based analyses traditionally used. But contemporary applications of land cover maps place greater demands on accuracy assessment, and sampling designs must be constructed to target objectives such as accuracy of land cover composition and landscape pattern. Sampling designs differ in their suitability to achieve different objectives, and trade-offs among desirable sampling design criteria must be recognized and accommodated when selecting a design. An overview is presented of the sampling designs used in accuracy assessment, and the status of these designs is appraised for meeting current needs. Sampling design features that facilitate multiple-objective accuracy assessments are described. |
| [18] | Accuracy assessment is a standard protocol of National Land Cover Database (NLCD) mapping. Here we report agreement statistics between map and reference labels for NLCD 2011, which includes land cover for ca. 2001, ca. 2006, and ca. 2011. The two main objectives were assessment of agreement between map and reference labels for the three, single-date NLCD land cover products at Level II and Level I of the classification hierarchy, and agreement for 17 land cover change reporting themes based on Level I classes (e.g., forest loss; forest gain; forest, no change) for three change periods (2001鈥2006, 2006 2011, and 2001 2011). The single-date overall accuracies were 82%, 83%, and 83% at Level II and 88%, 89%, and 89% at Level I for 2011, 2006, and 2001, respectively. Many class-specific user's accuracies met or exceeded a previously established nominal accuracy benchmark of 85%. Overall accuracies for 2006 and 2001 land cover components of NLCD 2011 were approximately 4% higher (at Level II and Level I) than the overall accuracies for the same components of NLCD 2006. The high Level I overall, user's, and producer's accuracies for the single-date eras in NLCD 2011 did not translate into high class-specific user's and producer's accuracies for many of the 17 change reporting themes. User's accuracies were high for the no change reporting themes, commonly exceeding 85%, but were typically much lower for the reporting themes that represented change. Only forest loss, forest gain, and urban gain had user's accuracies that exceeded 70%. Lower user's accuracies for the other change reporting themes may be attributable to the difficulty in determining the context of grass (e.g., open urban, grassland, agriculture) and between the components of the forest-shrubland-grassland gradient at either the mapping phase, reference label assignment phase, or both. NLCD 2011 user's accuracies for forest loss, forest gain, and urban gain compare favorably with results from other land cover change accuracy assessments. |
| [19] | The Himalaya, the world's highest mountain ranges, are home to a large group of glaciers and glacial lakes. Glacial lake outburst floods (GLOFs) in this region have resulted in catastrophic damages and fatalities in the past decades. The recent warming has caused dramatic glacial lake changes and increased potential GLOF risk in the Himalaya. However, our knowledge on the current state and change of glacial lakes in the entire Himalaya is limited. This study maps the current (2015) distribution of glacial lakes across the entire Himalaya and monitors the spatially-explicit evolution of glacial lakes over five time periods from 1990 to 2015 using a total of 348 Landsat images at 3002m resolution. The results show that 4950 glacial lakes in 2015 cover a total area of 455.302±0272.702km 2 , mainly located between 400002m and 570002m above sea level. Himalayan glacial lakes expanded by approximately 14.1% from 1990 to 2015. The changing patterns of supraglacial lakes and proglacial lakes are rather complex, involving both lake disappearance and emergence. Many emergent glacial lakes are found at higher elevations, especially the new proglacial lakes, which have formed as a result of glacier retreat. Spatially heterogeneous changes of Himalayan glacial lakes are observed, with the most significant expansion occurring in the southern slopes of the central Himalaya. Increasing glacier meltwater induced by the Himalayan atmospheric warming is a primary cause for the observed lake expansion. This study provides primary data for future GLOF risk assessments. A total of 118 rapidly expanded glacial lakes are identified as potential vulnerable lakes for the priority of risk assessment. |
| [20] | |
| [21] | . 在20世纪70年代中后期MSS图像、90年代初期TM图像和2004年TM/ETM图像支持下,通过三期遥感影像的直接对比分析,获得了三江源地区草地退化空间数据集,并在此基础上分析了70年代以来青海三江源地区草地退化的主要时空特征。结果表明:三江源地区草地退化是一个在空间格局上影响范围大,在时间过程上持续时间长的连续变化过程。研究发现,三江源草地退化的格局在70年代中后期已基本形成,70年代中后期至今,草地的退化过程一直在继续发生,总体上不存在90年代至今的草地退化急剧加强现象。草地退化的过程在不同区域和地带有明显不同的表现,如在湿润半湿润地带的草甸类草地上,发生着草地破碎化先导,随后发生覆盖度持续降低,最后形成黑土滩的退化过程;在干旱、半干旱地带的草原类草地上,发生着覆盖度持续降低,最后形成沙地和荒漠化草地的退化过程。三江源地区草地退化具有明显的区域差异,草地退化可以分为7个区,各区草地退化在类型、程度、范围与时间过程方面具有明显不同的特点。 在20世纪70年代中后期MSS图像、90年代初期TM图像和2004年TM/ETM图像支持下,通过三期遥感影像的直接对比分析,获得了三江源地区草地退化空间数据集,并在此基础上分析了70年代以来青海三江源地区草地退化的主要时空特征。结果表明:三江源地区草地退化是一个在空间格局上影响范围大,在时间过程上持续时间长的连续变化过程。研究发现,三江源草地退化的格局在70年代中后期已基本形成,70年代中后期至今,草地的退化过程一直在继续发生,总体上不存在90年代至今的草地退化急剧加强现象。草地退化的过程在不同区域和地带有明显不同的表现,如在湿润半湿润地带的草甸类草地上,发生着草地破碎化先导,随后发生覆盖度持续降低,最后形成黑土滩的退化过程;在干旱、半干旱地带的草原类草地上,发生着覆盖度持续降低,最后形成沙地和荒漠化草地的退化过程。三江源地区草地退化具有明显的区域差异,草地退化可以分为7个区,各区草地退化在类型、程度、范围与时间过程方面具有明显不同的特点。 |
| [22] | . 利用黄河源地区1985年和2000年1:100000土地利用/覆被数据,结合1:250000DEM、道路和居民点数据与野外调查资料,分析草地退化与坡向、海拔及距道路和居民点距离之间的关系,探讨黄河源区15年间土地覆被变化特征与规律。结果表明,退化草地占源区总面积的8.24%,冬春季牧场退化率显著高于夏季牧场;草地退化是黄河源区研究时段内土地利用/覆被变化最主要的特征。草地退化表现为:①阳坡退化率高于阴坡;②受人口密度影响,草地退化率与海拔高度成反比,相关系数为-0.925;③距离居民点越近,退化率越高。尤其当与居民点距离≤12km时,草地退化率与其相关系数高达-0.996;④在距道路4km以内,草地退化率与道路距离成反比,相关系数高达-0.978。1985年以来,源区的草地退化有自然因素的影响,但人类活动的影响仍起主导作用。科学地减少当地居民对草地的过分依赖是解决脆弱的江河源区环境退化的根本。 利用黄河源地区1985年和2000年1:100000土地利用/覆被数据,结合1:250000DEM、道路和居民点数据与野外调查资料,分析草地退化与坡向、海拔及距道路和居民点距离之间的关系,探讨黄河源区15年间土地覆被变化特征与规律。结果表明,退化草地占源区总面积的8.24%,冬春季牧场退化率显著高于夏季牧场;草地退化是黄河源区研究时段内土地利用/覆被变化最主要的特征。草地退化表现为:①阳坡退化率高于阴坡;②受人口密度影响,草地退化率与海拔高度成反比,相关系数为-0.925;③距离居民点越近,退化率越高。尤其当与居民点距离≤12km时,草地退化率与其相关系数高达-0.996;④在距道路4km以内,草地退化率与道路距离成反比,相关系数高达-0.978。1985年以来,源区的草地退化有自然因素的影响,但人类活动的影响仍起主导作用。科学地减少当地居民对草地的过分依赖是解决脆弱的江河源区环境退化的根本。 |
| [23] | . 青藏高原因强烈隆升，导致大气环流发生变化，形成了高原区域性环流。受此影响，从喀喇昆仑山最干旱的中段北翼的河尾滩、阿克赛钦一带向东延伸至中昆仑山南翼的黑石北湖、羊湖、白戈壁和昂歌库勒，及其以东的干旱区域，远离青藏高原两条水汽输送路径，气候寒冷干旱，成为亚洲的寒旱核心区域。据已有资料，寒旱核心区海拔在4700～5200m间，最暖月均温3～6℃，年降水量约为20～40mm。 本文在大量野外工作的基础上，基于改则及日土县境内19套土壤温度监测站点及3台自动气象站点1cm、10cm及20cm深度处土壤温度数据，分析了不同深度处土壤温度的季节变化和日变化;利... 青藏高原因强烈隆升，导致大气环流发生变化，形成了高原区域性环流。受此影响，从喀喇昆仑山最干旱的中段北翼的河尾滩、阿克赛钦一带向东延伸至中昆仑山南翼的黑石北湖、羊湖、白戈壁和昂歌库勒，及其以东的干旱区域，远离青藏高原两条水汽输送路径，气候寒冷干旱，成为亚洲的寒旱核心区域。据已有资料，寒旱核心区海拔在4700～5200m间，最暖月均温3～6℃，年降水量约为20～40mm。 本文在大量野外工作的基础上，基于改则及日土县境内19套土壤温度监测站点及3台自动气象站点1cm、10cm及20cm深度处土壤温度数据，分析了不同深度处土壤温度的季节变化和日变化;利... |
| [24] | . <p>青藏高原独特而敏感的生态系统，是中国生态安全屏障的重要载体，也是区域经济发展的必要基础。生态建<br />设是生态文明的需求，也是社会经济可持续发展的保障。由于青藏高原内部自然条件差异显著，各区生态功能及区<br />域问题也不尽相同，生态建设亟需因地制宜地提出建设与保护的对策和措施。本文从生态系统的主要功能、脆弱程<br />度、变化趋势及面临风险等特征，辨识出阿里西部、那曲中南部、三江源地区和三江并流区等4 个生态建设的关键<br />区；并在分析各区环境和生态特征与土地覆被变化及其原因的基础上，提出了稳定和提升青藏高原生态安全屏障功<br />能的措施与建议，对于高原生态安全屏障保护与建设具有重要的参考价值。</p> <p>青藏高原独特而敏感的生态系统，是中国生态安全屏障的重要载体，也是区域经济发展的必要基础。生态建<br />设是生态文明的需求，也是社会经济可持续发展的保障。由于青藏高原内部自然条件差异显著，各区生态功能及区<br />域问题也不尽相同，生态建设亟需因地制宜地提出建设与保护的对策和措施。本文从生态系统的主要功能、脆弱程<br />度、变化趋势及面临风险等特征，辨识出阿里西部、那曲中南部、三江源地区和三江并流区等4 个生态建设的关键<br />区；并在分析各区环境和生态特征与土地覆被变化及其原因的基础上，提出了稳定和提升青藏高原生态安全屏障功<br />能的措施与建议，对于高原生态安全屏障保护与建设具有重要的参考价值。</p> |
| [25] | |
| [26] | A new global land cover database for the year 2000 (GLC2000) has been produced by an international partnership of 30 research groups coordinated by the European Commission's Joint Research Centre. The database contains two levels of land cover information—detailed, regionally optimized land cover legends for each continent and a less thematically detailed global legend that harmonizes regional legends into one consistent product. The land cover maps are all based on daily data from the VEGETATION sensor on‐board SPOT 4, though mapping of some regions involved use of data from other Earth observing sensors to resolve specific issues. Detailed legend definition, image classification and map quality assurance were carried out region by region. The global product was made through aggregation of these. The database is designed to serve users from science programmes, policy makers, environmental convention secretariats, non‐governmental organizations and development‐aid projects. The regional and global data are available free of charge for all non‐commercial applications from http://www.gvm.jrc.it/glc2000. |
| [27] | Researchers from the U.S. Geological Survey, University of Nebraska-Lincoln and the European Commission''s Joint Research Centre, Ispra, Italy produced a 1 km resolution global land cover characteristics database for use in a wide range of continental-to global-scale environmental studies. This database provides a unique view of the broad patterns of the biogeographical and ecoclimatic diversity of the global land surface, and presents a detailed interpretation of the extent of human development. The project was carried out as an International Geosphere-Biosphere Programme, Data and Information Systems (IGBP-DIS) initiative. The IGBP DISCover global land cover product is an integral component of the global land cover database. DISCover includes 17 general land cover classes defined to meet the needs of IGBP core science projects. A formal accuracy assessment of the DISCover data layer will be completed in 1998. The 1 km global land cover database was developed through a continent-by-continent unsupervised classification of 1 km monthly Advanced Very High Resolution Radiometer (AVHRR) Normalized Difference Vegetation Index (NDVI) composites covering 1992-1993. Extensive post-classification stratification was necessary to resolve spectral/temporal confusion between disparate land cover types. The complete global database consists of 961 seasonal land cover regions that capture patterns of land cover, seasonality and relative primary productivity. The seasonal land cover regions were aggregated to produce seven separate land cover data sets used for global environmental modelling and assessment. The data sets include IGBP DISCover, U.S. Geological Survey Anderson System, Simple Biosphere Model, Simple Biosphere Model 2, Biosphere-Atmosphere Transfer Scheme, Olson Ecosystems and Running Global Remote Sensing Land Cover. The database also includes all digital sources that were used in the classification. The complete database can be sourced from the website: http://edcwww.cr.usgs.gov/landdaac/glcc/glcc.html. |
| [28] | This paper on reports the production of a 1 km spatial resolution land cover classification using data for 1992-1993 from the Advanced Very High Resolution Radiometer (AVHRR). This map will be included as an at-launch product of the Moderate Resolution Imaging Spectroradiometer (MODIS) to serve as an input for several algorithms requiring knowledge of land cover type. The methodology was derived from a similar effort to create a product at 8 km spatial resolution, where high resolution data sets were interpreted in order to derive a coarse-resolution training data set. A set of 37 294 x 1 km pixels was used within a hierarchical tree structure to classify the AVHRR data into 12 classes. The approach taken involved a hierarchy of pair-wise class trees where a logic based on vegetation form was applied until all classes were depicted. Multitemporal AVHRR metrics were used to predict class memberships. Minimum annual red reflectance, peak annual Normalized Difference Vegetation Index (NDVI), and minimum channel three brightness temperature were among the most used metrics. Depictions of forests and woodlands, and areas of mechanized agriculture are in general agreement with other sources of information, while classes such as low biomass agriculture and high-latitude broadleaf forest are not. Comparisons of the final product with regional digital land cover maps derived from high-resolution remotely sensed data reveal general agreement, except for apparently poor depictions of temperate pastures within areas of agriculture. Distinguishing between forest and non-forest was achieved with agreements ranging from 81 to 92% for these regional subsets. The agreements for all classes varied from an average of 65% when viewing all pixels to an average of 82% when viewing only those 1 km pixels consisting of greater than 90% one class within the high-resolution data sets. |
| [29] | Information related to land cover is immensely important to global change science. In the past decade, data sources and methodologies for creating global land cover maps from remote sensing have evolved rapidly. Here we describe the datasets and algorithms used to create the Collection 5 MODIS Global Land Cover Type product, which is substantially changed relative to Collection 4. In addition to using updated input data, the algorithm and ancillary datasets used to produce the product have been refined. Most importantly, the Collection 5 product is generated at 500-m spatial resolution, providing a four-fold increase in spatial resolution relative to the previous version. In addition, many components of the classification algorithm have been changed. The training site database has been revised, land surface temperature is now included as an input feature, and ancillary datasets used in post-processing of ensemble decision tree results have been updated. Further, methods used to correct classifier results for bias imposed by training data properties have been refined, techniques used to fuse ancillary data based on spatially varying prior probabilities have been revised, and a variety of methods have been developed to address limitations of the algorithm for the urban, wetland, and deciduous needleleaf classes. Finally, techniques used to stabilize classification results across years have been developed and implemented to reduce year-to-year variation in land cover labels not associated with land cover change. Results from a cross-validation analysis indicate that the overall accuracy of the product is about 75% correctly classified, but that the range in class-specific accuracies is large. Comparison of Collection 5 maps with Collection 4 results show substantial differences arising from increased spatial resolution and changes in the input data and classification algorithm. |
| [30] | |
| [31] | [Excerpt: Introduction] In 2008, the ESA-GlobCover 2005 project delivered to the international community the very first 300- m global land cover map for 2005 as well as bimonthly and annual MERIS (Medium Resolution Imaging Spectrometer Instrument) Fine Resolution (FR) surface reflectance mosaics. The ESA- GlobCover 2005 project, carried out by an international consortium, started in April 2005 and relied on very rich feedback and comments from a large partnership including end-users belonging to international institutions (JRC, FAO, EEA, UNEP, GOFC-GOLD and IGBP) in addition to ESA internal... |
| [32] | |
| [33] | Over 10,000 Landsat-like satellite images are required to cover the entire Earth at 30 m resolution. To derive a GLC map from such a large volume of data necessitates the development of effective, efficient, economic and operational approaches. Automated approaches usually provide higher efficiency and thus more economic solutions, yet existing automated classification has been deemed ineffective because of the low classification accuracy achievable (typically below 65%) at global scale at 30 m resolution. As a result, an approach based on the integration of pixel- and object-based methods with knowledge (POK-based) has been developed. To handle the classification process of 10 land cover types, a split-and-merge strategy was employed, i.e. firstly each class identified in a prioritized sequence and then results are merged together. For the identification of each class, a robust integration of pixel-and object-based classification was developed. To improve the quality of the classification results, a knowledge-based interactive verification procedure was developed with the support of web service technology. The performance of the POK-based approach was tested using eight selected areas with differing landscapes from five different continents. An overall classification accuracy of over 80% was achieved. This indicates that the developed POK-based approach is effective and feasible for operational GLC mapping at 30 m resolution. |
| [34] | Accurate and up-to-date global land cover data sets are necessary for various global change research studies including climate change, biodiversity conservation, ecosystem assessment, and environmental modeling. In recent years, substantial advancement has been achieved in generating such data products. Yet, we are far from producing geospatially consistent high-quality data at an operational level. We compared the recently available Global Land Cover 2000 (GLC-2000) and MODerate resolution Imaging Spectrometer (MODIS) global land cover data to evaluate the similarities and differences in methodologies and results, and to identify areas of spatial agreement and disagreement. These two global land cover data sets were prepared using different data sources, classification systems, and methodologies, but using the same spatial resolution (i.e., 1 km) satellite data. Our analysis shows a general agreement at the class aggregate level except for savannas/shrublands, and wetlands. The disagreement, however, increases when comparing detailed land cover classes. Similarly, percent agreement between the two data sets was found to be highly variable among biomes. The identified areas of spatial agreement and disagreement will be useful for both data producers and users. Data producers may use the areas of spatial agreement for training area selection and pay special attention to areas of disagreement for further improvement in future land cover characterization and mapping. Users can conveniently use the findings in the areas of agreement, whereas users might need to verify the informaiton in the areas of disagreement with the help of secondary information. Learning from past experience and building on the existing infrastructure (e.g., regional networks), further research is necessary to (1) reduce ambiguity in land cover definitions, (2) increase availability of improved spatial, spectral, radiometric, and geometric resolution satellite data, and (3) develop advanced classification algorithms. |
| [35] | Precise global/regional land cover mapping is foundational data in the studies of land surface processes and modelling.The quantitative assessments of the map quality and classification accuracy for the available land cover maps will help to improve the accuracy of land cover mapping in future.Four land cover data sets over China are compared and evaluated.The data sets include Version 2 global land cover dataset of IGBP,the MODIS Land cover map 2001,global land cover map produced by the University of Maryland,and the land cover map from Global Vegetation Monitoring unit of the European Commission Joint Research Centre(GLC2000).The four maps have different classification systems,which makes the comparison difficult.It is necessary to aggregate these maps by reclassifying them using a unified legend system.A large scale,i.e.1 100 000 land cover map of China was used as the reference data to validate the four maps.The results show that the GLC2000 land cover map has the highest accuracy.But it has obvious rating error locally and its rating accuracy is zero for wetland.The MODIS land cover map ranks second for type area consistency and ranks third for sub fraction overall accuracy as compared with reference data,which may be affect by local rating error.The IGBP land cover map hast good rating accuracy though it has a local rating error and third consistency for type area.The rating accuracy and type area consistency with reference data of UMD land cover map is low wholly.It is believed that the accuracies of all the datasets can not meet the requirement of land surface modelling.Addition,an information fusion strategy is proposed to produce a higher accuracy land cover map of China,whose classification system should be compatible with the well-accepted classification system used in land surface modelling. |
| [36] | Accurate and up-to-date global land cover data sets are necessary for various global change research studies including climate change, biodiversity conservation, ecosystem assessment, and environmental modeling. In recent years, substantial advancement has been achieved in generating such data products. Yet, we are far from producing geospatially consistent high-quality data at an operational level. We compared the recently available Global Land Cover 2000 (GLC-2000) and MODerate resolution Imaging Spectrometer (MODIS) global land cover data to evaluate the similarities and differences in methodologies and results, and to identify areas of spatial agreement and disagreement. These two global land cover data sets were prepared using different data sources, classification systems, and methodologies, but using the same spatial resolution (i.e., 1 km) satellite data. Our analysis shows a general agreement at the class aggregate level except for savannas/shrublands, and wetlands. The disagreement, however, increases when comparing detailed land cover classes. Similarly, percent agreement between the two data sets was found to be highly variable among biomes. The identified areas of spatial agreement and disagreement will be useful for both data producers and users. Data producers may use the areas of spatial agreement for training area selection and pay special attention to areas of disagreement for further improvement in future land cover characterization and mapping. Users can conveniently use the findings in the areas of agreement, whereas users might need to verify the informaiton in the areas of disagreement with the help of secondary information. Learning from past experience and building on the existing infrastructure (e.g., regional networks), further research is necessary to (1) reduce ambiguity in land cover definitions, (2) increase availability of improved spatial, spectral, radiometric, and geometric resolution satellite data, and (3) develop advanced classification algorithms. |
| [37] | Precise global/regional land cover mapping is foundational data in the studies of land surface processes and modelling.The quantitative assessments of the map quality and classification accuracy for the available land cover maps will help to improve the accuracy of land cover mapping in future.Four land cover data sets over China are compared and evaluated.The data sets include Version 2 global land cover dataset of IGBP,the MODIS Land cover map 2001,global land cover map produced by the University of Maryland,and the land cover map from Global Vegetation Monitoring unit of the European Commission Joint Research Centre(GLC2000).The four maps have different classification systems,which makes the comparison difficult.It is necessary to aggregate these maps by reclassifying them using a unified legend system.A large scale,i.e.1鈭100 000 land cover map of China was used as the reference data to validate the four maps.The results show that the GLC2000 land cover map has the highest accuracy.But it has obvious rating error locally and its rating accuracy is zero for wetland.The MODIS land cover map ranks second for type area consistency and ranks third for sub fraction overall accuracy as compared with reference data,which may be affect by local rating error.The IGBP land cover map hast good rating accuracy though it has a local rating error and third consistency for type area.The rating accuracy and type area consistency with reference data of UMD land cover map is low wholly.It is believed that the accuracies of all the datasets can not meet the requirement of land surface modelling.Addition,an information fusion strategy is proposed to produce a higher accuracy land cover map of China,whose classification system should be compatible with the well-accepted classification system used in land surface modelling. |
| [38] | 61Provides good practice recommendations for accuracy assessment and area estimation61Recommendations are a synthesis of existing methods.61Recommendations satisfy accepted scientific practice.61Recommendations address sampling design, response design and analysis.61An example illustrating recommended workflow is included. |
| [39] | . 以河南省为研究区,对全球首套30 m分辨率土地覆盖产品Globle Land30进行区域尺度精度评价。首先,以中国1∶10万土地利用数据(CHINA-2010)为参考,分析两种产品的空间一致性;而后,通过Google Earth样本分析Globle Land30在空间不一致区域的制图精度;最后,利用野外实地考察样本对Globle Land30进行总体精度评价,并从土地覆被复杂度、高程等方面分析影响精度的原因,结果表明:1 Globle Land30与CHINA-2010空间一致性达80.20%。两种产品对耕地、林地、人工地面一致性高,对草地、水体、灌木、湿地、未利用土地的一致性低。2在空间不一致区域,Globle Land30的总体分类正确率略低于CHINA-2010,但两者对不同地类的优势不同。3经野外实地考察验证可知,Globle Land30的总体精度达83.33%。4GlobleLand30与CHINA-2010的空间一致性随土地覆被复杂度的增加而降低,并在高程过渡带较低。 以河南省为研究区,对全球首套30 m分辨率土地覆盖产品Globle Land30进行区域尺度精度评价。首先,以中国1∶10万土地利用数据(CHINA-2010)为参考,分析两种产品的空间一致性;而后,通过Google Earth样本分析Globle Land30在空间不一致区域的制图精度;最后,利用野外实地考察样本对Globle Land30进行总体精度评价,并从土地覆被复杂度、高程等方面分析影响精度的原因,结果表明:1 Globle Land30与CHINA-2010空间一致性达80.20%。两种产品对耕地、林地、人工地面一致性高,对草地、水体、灌木、湿地、未利用土地的一致性低。2在空间不一致区域,Globle Land30的总体分类正确率略低于CHINA-2010,但两者对不同地类的优势不同。3经野外实地考察验证可知,Globle Land30的总体精度达83.33%。4GlobleLand30与CHINA-2010的空间一致性随土地覆被复杂度的增加而降低,并在高程过渡带较低。 |
| [40] | |
| [41] | |
| [42] | Asia’s high plateaus are sensitive to climate change and have been experiencing rapid warming over the past few decades. We found 99 new lakes and extensive lake expansion on the Tibetan Plateau during the last four decades, 1970–2013, due to increased precipitation and cryospheric contributions to its water balance. This contrasts with disappearing lakes and drastic shrinkage of lake areas on the adjacent Mongolian Plateau: 208 lakes disappeared, and 75% of the remaining lakes have shrunk. We detected a statistically significant coincidental timing of lake area changes in both plateaus, associated with the climate regime shift that occurred during 1997/1998. This distinct change in 1997/1998 is thought to be driven by large-scale atmospheric circulation changes in response to climate warming. Our findings reveal that these two adjacent plateaus have been changing in opposite directions in response to climate change. These findings shed light on the complex role of the regional climate and water cycles and provide useful information for ecological and water resource planning in these fragile landscapes. |
| [43] | . In this study, using the common classification systems of IGBP-17, IGBP-9, IPCC-5 and TC (vegetation, wetlands and others only), we studied spatial and areal inconsistencies in the three most recent multi-resource land cover products in a complex mountain-oasis-desert system and quantitatively discussed the uncertainties in classification system conversion. This is the first study to compare these products based on terrain and to quantitatively study the uncertainties in classification system conversion. The inconsistencies and uncertainties decreased from high to low levels of aggregation (IGBP-17 to TC) and from mountain to desert areas, indicating that the inconsistencies are not only influenced by the level of thematic detail and landscape complexity but also related to the conversion uncertainties. The overall areal inconsistency in the comparison of the FROM-GLC and GlobCover 2009 datasets is the smallest among the three pairs, but the smallest overall spatial inconsistency was observed between the FROM-GLC and MODISLC. The GlobCover 2009 had the largest conversion uncertainties due to mosaic land cover definition, with values up to 23.9%, 9.68% and 0.11% in mountainous, oasis and desert areas, respectively. The FROM-GLC had the smallest inconsistency, with values less than 4.58%, 1.89% and 1.2% in corresponding areas. Because the FROM-GLC dataset uses a hierarchical classification scheme with explicit attribution from the second level to the first, this system is suggested for producers of map land cover products in the future. |
| [44] | . A detailed analysis was made of the response of olive floral phenology to climate and topography in southern Spain. Field phenological, topographical and meteorological data collected at 12 sampling sites in the province of Cordoba over a 17-year period (1996-2012) were statistically analyzed and used to model local olive phenological behaviour.The study sought to determine: (1) the optimal frequency of phenological sampling during the reproductive period; (2) the major topographical parameters governing local olive reproductive phenology; and (3) the most influential meteorological variables. Findings for the Sign test indicated that weekly sampling yielded accurate results. Correlation and multiple linear regression analysis revealed that altitude and percentage eastward slope were the most influential topographical factors; a positive correlation was detected between delays in phenophases onset and increased altitude and eastward orientation. Correlation and partial least square regression analysis identified air temperature, rainfall, crop evapotranspiration and solar radiation as the major weather factors influencing local olive phenology. (C) 2013 Elsevier B.V. All rights reserved. |
| [45] | As the Earth's third pole, the Tibetan Plateau has experienced a pronounced warming in the past decades. Recent studies reported that the start of the vegetation growing season (SOS) in the Plateau showed an advancing trend from 1982 to the late 1990s and a delay from the late 1990s to 2006. However, the findings regarding the SOS delay in the later period have been questioned, and the reasons causing the delay remain unknown. Here we explored the alpine vegetation SOS in the Plateau from 1982 to 2011 by integrating three long-term time-series datasets of Normalized Difference Vegetation Index (NDVI): Global Inventory Modeling and Mapping Studies (GIMMS, 1982-2006), SPOT VEGETATION (SPOT-VGT, 1998-2011), and Moderate Resolution Imaging Spectroradiometer (MODIS, 2000-2011). We found GIMMS NDVI in 2001-2006 differed substantially from SPOT-VGT and MODIS NDVIs and may have severe data quality issues in most parts of the western Plateau. By merging GIMMS-based SOSs from 1982 to 2000 with SPOT-VGT-based SOSs from 2001 to 2011 we found the alpine vegetation SOS in the Plateau experienced a continuous advancing trend at a rate of similar to 1.04 d.y(-1) from 1982 to 2011, which was consistent with observed warming in springs and winters. The satellite-derived SOSs were proven to be reliable with observed phenology data at 18 sites from 2003 to 2011; however, comparison of their trends was inconclusive due to the limited temporal coverage of the observed data. Longer-term observed data are still needed to validate the phenology trend in the future. |
