| 基于NDVI的新疆荒漠地区植被覆盖度遥感估算经验模型研究 |
| |
| 引用本文： | 岳健,穆桂金,唐自华,杨雪峰,林永崇,徐立帅. 基于NDVI的新疆荒漠地区植被覆盖度遥感估算经验模型研究[J]. 干旱区地理, 2020, 43(1): 153-160. DOI: 10.12118/j.issn.1000-6060.2020.01.18 |
| |
| 作者姓名： | 岳健 穆桂金 唐自华 杨雪峰 林永崇 徐立帅 |
| |
| 作者单位： | 1中国科学院新疆生态与地理研究所荒漠与绿洲生态国家重点实验室，新疆乌鲁木齐830011； 2新疆策勒荒漠草地生态系统国家野外科学观测研究站，新疆策勒848300； 3中国科学院地质与地球物理研究所，北京100029；4新疆师范大学地理科学与旅游学院，新疆乌鲁木齐830054； 5闽南师范大学历史地理学院，福建漳州363000；6山西农业大学资源与环境学院，山西太古030801 |
| |
| 基金项目： | 新疆维吾尔自治区自然科学基金项目（2015211A049） |
| |
| 摘 要： | 新疆荒漠地区植被覆盖度遥感估算模型十分缺乏,给荒漠化监测等相关工作带来很大不便,开展植被覆盖度遥感估算经验模型研究,对于促进和完善相关地区的生态监测及研究工作具有积极的现实意义。通过对阜康市北部沙漠南缘和克拉玛依市中部平原荒漠进行无人机航拍,利用无人机遥感提取(光合)植被信息,并将无人机航拍影像的植被覆盖度统计单元与高分辨率卫星影像像元在空间上直接相对应,获取在高分辨率卫星影像像元尺度上的植被盖度,然后通过植被覆盖度和空间上与其相对应的源自高分辨率卫星影像的NDVI数据的拟合关系,建立基于源自高分二号影像的NDVI的阜康北部沙漠植被覆盖度遥感估算线性模型以及基于源自ZY1-02C影像的NDVI的克拉玛依平原荒漠植被覆盖度遥感估算二次多项式模型。研究中所采用的无人机遥感与卫星遥感相结合、植被覆盖度统计单元与卫星像元在空间上直接对应的方法,可避免以往相关工作中常以点位测量数据代表卫星像元数据所带来的不确定性。由于所用卫星影像的NDVI数据稳定性相对不足等原因,所建立的遥感估算模型的估算精度尚相对偏低,有待于今后进一步的工作加以改进。
|
| 关 键 词： | NDVI 荒漠 植被覆盖度 遥感估算 经验模型 |
| 收稿时间： | 2019-05-11 |
Remote sensing estimation models for vegetation coverage in desert regions of Xinjiang based on NDVI |
| |
| YUE Jian,MU Gui-jin,TANG Zi-hua,YANG Xue-feng,LIN Yong-chong,XU Li-shuai. Remote sensing estimation models for vegetation coverage in desert regions of Xinjiang based on NDVI[J]. Arid Land Geography, 2020, 43(1): 153-160. DOI: 10.12118/j.issn.1000-6060.2020.01.18 |
| |
| Authors: | YUE Jian MU Gui-jin TANG Zi-hua YANG Xue-feng LIN Yong-chong XU Li-shuai |
| |
| Abstract: | The lack of remote sensingestimation models in vegetation coverage for desert regions of Xinjiang, China has brought great inconvenience todesertification monitoring and other related work. It is of positive and practical significance to studythe empirical models for remote sensing estimation of vegetation coverage forpromoting and improving ecological monitoring abilities in relevant areas. In this study, unmanned aerial vehicle (UAV) photography was firstlycarried out in typical desert regions in Xinjiang, vegetation information (photosynthesis) was extractedby UAV images. Secondly, vegetation coverage was obtained at the pixel scale ofhigh resolution satellite image by making the statistical units of vegetationcoverage directly correspond to the pixels of high resolution satellite imagein space. Lastly, based on the fitting relationship between vegetationcoverage and the corresponding NDVI data derived fromhigh resolution satellite images, the empirical modelswere established for estimating vegetation coverage of typical desert regionsin Xinjiang. The deserts in thesouthern margin of the northern part of Fukang City and in the central plain ofKaramay City were chosen as aerial shooting areas respectively. The relative flight height was set to 6-7 m and theground resolution was 0.002-0.003 m. Satellite images which were shot on the same day andthe day before UAV photographing, were selected andpurchased for generating NDVI data. Agisoft Photoscan software was used for UAV imageprocessing, ENVI and ArcGISsoftware packages were used for satellite remote sensing image processing andspatial analysis, Origin software wasused for fitting analysis of data series and the vegetation information of UAVimages was extracted by NDVI index. As a result, a linear model (y=149.86x-13.449,R2=0.735 3) was established for remote sensingestimation of vegetation coverage in northern sandy desert in Fukang based on NDVI derived from GF2 satellite imageand a quadratic polynomial model （y=97.397x2+80.837x-5.210 9,R2=0.818） was established for remote sensing estimation ofvegetation coverage in plain desert in Karamay based on NDVI derived from ZY1-02C satellite image. These models might provide necessary foundation andbasis for monitoring and research of land use and land cover change, ecological environment change, desertification (or sandy desertification),vegetation mapping and otherrelated work in Xinjiang. The combination ofUAV images and satellite remotely sensed data and the spatial directcorrespondence between vegetation coverage statistical unit and satellitepixel, could avoid the uncertainty caused by the process ofrepresenting vegetation coverage of satellite pixel with manual measurementdata at different points in the previous related work. The proposed method made the spatial matching betweenvegetation coverage and satellite image pixels relatively more intuitive andaccurate in the process of constructing remote sensing estimation model ofvegetation coverage and it also provided an idea for similar work. |
| |
| Keywords: | NDVI desert vegetation coverage remote sensing estimation empirical model |
|
| 点击此处可从《干旱区地理》浏览原始摘要信息 |
|
点击此处可从《干旱区地理》下载免费的PDF全文 |
|
