自然资源学报 ›› 2015, Vol. 30 ›› Issue (2): 260-270.doi: 10.11849/zrzyxb.2015.02.009

• 资源评价 • 上一篇    下一篇

基于TRMM订正数据的横断山区降水时空分布特征

张涛1,2,3, 李宝林1, 何元庆2, 杜建括2,3, 牛贺文2, 辛惠娟2,3   

  1. 1. 中国科学院地理科学与资源研究所资源与环境信息系统国家重点实验室, 北京100101;
    2. 中国科学院寒区旱区环境与工程研究所冰冻圈科学国家重点实验室, 兰州730000;
    3. 中国科学院大学, 北京100049
  • 收稿日期:2013-12-24 修回日期:2014-06-20 出版日期:2015-02-20 发布日期:2015-02-10
  • 通讯作者: 李宝林(1970-),男,博士,研究员,主要从事遥感环境变化检测研究。E-mail:libl@lreis.ac.cn E-mail:libl@lreis.ac.cn
  • 作者简介:张涛(1987-),男,四川泸县人,博士研究生,主要从事环境遥感研究。E-mail:zhangtao@lreis.ac.cn
  • 基金资助:

    973 项目(2015CB954101,2015CB954103);科技支撑计划项目(2012BAH33B01)。

Spatial and Temporal Distribution of Precipitation Based on Corrected TRMM Data in Hengduan Mountains

ZHANG Tao1,2,3, LI Bao-lin1, HE Yuan-qing2, DU Jian-kuo2,3, NIU He-wen2, XIN Hui-juan2,3   

  1. 1. State Key Laboratory of Resources and Environmental Information System, Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China;
    2. State Key Laboratory of Cryosphere Science, Cold and Arid Region Environment and Engineering Research Institute, CAS, Lanzhou 730000, China;
    3. University of Chinese Academy of Sciences, Beijing 100049, China
  • Received:2013-12-24 Revised:2014-06-20 Online:2015-02-20 Published:2015-02-10
  • About author:10.11849/zrzyxb.2015.02.009

摘要:

鉴于高时空变异地区降水观测的需求,论文提出了基于ISODATA动态聚类法和最大似然法分区逐月回归的TRMM数据订正方法,并以横断山区为例,利用地基台站降水数据对1998—2012 年的TRMM 3B43 V7 数据进行了实验研究,并探讨了过去15 a 横断山区的降水时空分布特征及变化趋势。结果表明,TRMM 3B43 V7 数据在横断山区的总体精度较高,单站精度较低。订正后的TRMM 3B43 数据,与实测值偏差大为减少,降水相对偏差大于10%的站点由原始数据的16 个(42.1%)减少为7 个(18.4%),有81.6%的站点年降水量相对偏差小于10%,且相对偏差大于等于20%的站点仅3 个,能够基本满足横断山区降水时空分布特征研究的精度需要,有效地弥补了有限站点观测的不足。研究区内年降水量从东南向西北递减,与东亚季风在该区域的走向一致;1998—2012 年横断山区的降水量主要呈减少趋势,减少地区主要分布在南部以及中西部地区。夏季降水减少趋势最为突出,秋季次之,冬春变化趋势不明显。横断山区大部分区域的降水量与东亚夏季风指数正相关,与东亚冬季风指数负相关;最近15 a,研究区东亚季风指数持续变小,恰好与研究结果中的降水减少趋势一致。

关键词: TRMM卫星, 降水, 数据订正, 横断山区

Abstract:

Precipitation measurement is always an important problem in mountainous areas. Because of multiple factors such as atmospheric circulation, topography, and land-sea positions, precipitation shows strong spatial-temporal heterogeneity, especially in mountainous areas. Meanwhile, although traditional rain gauges and rainfall measuring radar on the ground can obtain the precipitation information precisely, the distribution density and range of the observation sites are the main limitations. Tropical Rainfall Measuring Mission (TRMM) data may fill the gaps. The Hengduan Mountains, which is located in the southeast of Qinghai-Tibet plateau, has been chosen as the study area in this article. There are 38 weather stations interspersing in this region, and the recorded monthly precipitations from 1998 to 2012 have been used as the truth values, hence the TRMM 3B43 V7 data are verified during the same period. After that, a calibration method based on linear regression of monthly TRMM 3B43 data of each subregion has been carried out by using ISODATA clustering and maximum likelihood method. Finally, the spatial and temporal distribution features of precipitation in Hengduan Mountains are investigated. Results show that TRMM 3B43 data have significant correlation with perticipations, while have low precision at almost half of the sites. After correction, the precision is improved greatly. The number of the stations whose biases (compared with the rain gauges) are more than 10% decreased from 16 (42.1%) to 7 (18.4%), that is, 81.6% of the stations have the biases below 10%, and only 3 stations have TRMM biases ≥20%, which is acceptable in the study area. The precipitation decreases gradually from southeast to northwest, which shows the same tendency as the East Asian Monsoon in Hengduan Mountains. Precipitation in the study area is decreasing during 1998-2012, mainly in the southern and mid-western regions. Precipitation in summer displays the most decrease, that in autumn ranks the second, while there are no significant changes in spring and winter. Precipitation in most of the Hengduan Mountains shows positive correlation with the East Asian Summer Monsoon Index, while displays negative correlation with the East Asian Winter Monsoon Index. The East Asian Summer Monsoon Index in this area is continuously shrinking in the last 15 years, which is consistent with the decreasing trend of precipitation indicated by this article.

Key words: Hengduan Mountains, precipitation, TRMM satellite, data correction

中图分类号: 

  • P332.1