img

官方微信

img

群号:冰川冻土交流群

QQ群:218834310

高级检索

冰川冻土 ›› 2016, Vol. 38 ›› Issue (1): 159-169.doi: 10.7522/j.isnn.1000-0240.2016.0018

• 寒区科学与技术 • 上一篇    下一篇

西藏高原MODIS每日积雪产品去云算法过程对比验证研究

拉巴卓玛1,2,3,4, 邱玉宝2, 次旦巴桑5, 郑照军6, 拉巴4, 陈涛4, 张欢2, 除多4   

  1. 1. 中国气象局 成都高原气象研究所, 四川 成都 610072;
    2. 中国科学院 遥感与数字地球科学研究所, 北京 100094;
    3. 成都信息工程大学 资源环境学院, 四川 成都 610225;
    4. 西藏高原大气环境科学研究所, 西藏 拉萨 850000;
    5. 西藏自治区气象台, 西藏 拉萨 850000;
    6. 中国气象局 国家卫星气象中心, 北京 100081
  • 收稿日期:2015-11-10 修回日期:2016-01-09 出版日期:2016-02-25 发布日期:2016-05-30
  • 作者简介:拉巴卓玛(1984-),女,西藏拉萨人,工程师,2015年在成都信息工程大学获硕士学位,从事遥感应用研究.E-mail:lhakdron@126.com.
  • 基金资助:
    公益性行业(气象)科研专项(GYHY201206040);国家自然科学基金项目(4137135;41165002;41165003);中国气象局关键技术集成项目(CMAGJ2014M50)资助

The validation of MODIS daily snow-cover products after cloud removal in Tibet Autonomous Region

Laba Zhuoma1,2,3,4, QIU Yubao2, Cidan Basang5, ZHENG Zhaojun6, Laba4, CHEN Tao4, ZHANG Huan2, Chu Duo4   

  1. 1. Institute of Plateau Meteorology, China Meteorological Administration, Chengdu 610072, China;
    2. Key Laboratory of Digital Earth Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100094, China;
    3. Chengdu University of Information Technology, Chengdu 610225, China;
    4. Tibetan Institute of Atmospheric Environment and Science, Lhasa 850000, China;
    5. Tibet Meteorological Bureau, Lhasa 850000, China;
    6. National Satellite Meteorological Center, China Meteorological Administration, Beijing 100081, China
  • Received:2015-11-10 Revised:2016-01-09 Online:2016-02-25 Published:2016-05-30

摘要: 基于MODIS积雪产品的多种去云方法过程在西藏高原开展对比验证,并以漏测误差L(%)、多测误差M(%)、总体精度O(%)、积雪分类精度S(%)及kappa系数(Khat)作为验证精度评价指标;主要研究方法有上下午星结合去云法(方法Ⅰ)、连续三天结合去云法(方法Ⅱ)以及基于数字高程模型的Snowl去云法(方法Ⅲ).并采用高分辨率Landsat遥感数据对所得结果开展验证对比工作,为了使验证结果更具有代表性,选取了西藏高原典型的5个区域用于其验证范围.结果表明:方法Ⅰ和方法Ⅱ有较好的去云效果,适合于在西藏高原开展利用;方法Ⅲ在西藏高原没有起到特殊的去云作用,该方法的前提假设是在西藏高原需要考虑山地和高原作分析;通过三种方法去云之后在5个区域效果总体上表现为漏测误差和多测误差都有减少,总体精度和积雪分类精度分别在60%~98%及60%~95%之间,kappa系数在0.10~0.56间,一致性精度表现一般.

关键词: 积雪, 遥感, MODIS, 西藏, 去云

Abstract: In this paper, the method of cloud removal applied to MODIS/TERRA and AQUA snow products was validated in Tibet Autonomous Region, the leakage measurement error L (%), more measurement error M (%), overall accuracy O (%) and kappa coefficient (Khat) was chosen as evaluation index; cloud removal of combination of TERRA and AQUA products (Ⅰ), cloud removal of combination of snow products for three days (Ⅱ) and Snowl cloud removal based on DEM (Ⅲ) were the three methods of cloud removal work. In addition, five typical regions of Tibet were chosen to be researching places, so as to make the study more convincing. The results show that Methods (Ⅰ) and (Ⅱ) are suitable to use in this research areas for cloud removal, and there are many uncertainties in Method (Ⅲ). After cloud removal by the three methods, for the five regions, it is found that the overall accuracy O rising to 60%-98%, the snow classification precision S rising to 60%-95%, and the Kappa coefficient ranging from 0.10 to 0.56 in general.

Key words: snow cover, remote sensing, MODIS, Tibet, cloud removal

中图分类号: 

  • P426.63+5