| 中亚地区1982年至2002年植被指数与气温和降水的相关性分析 |
| |
| 引用本文： | 索玉霞,王正兴,刘 闯,于伯华.中亚地区1982年至2002年植被指数与气温和降水的相关性分析[J].资源科学,2009,31(8):1422-1429. |
| |
| 作者姓名： | 索玉霞 王正兴 刘 闯 于伯华 |
| |
| 作者单位： | 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;中国科学院研究生院,北京,100049
2. 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 |
| |
| 基金项目： | 资源与环境信息系统国家重点实验室自主创新项目 |
| |
| 摘 要： | 本文利用1982年～2002年间AVHRR-NDVI数据和气候研究组(CRU)降水与气温数据,分析了中亚5国21年来NDVI年际与季节变化特征及其与气候因子的相关关系.结果表明:①在植被生长季,53%地区NDVI年变化率＜±0.0005NDVI/a(无变化),40%地区NDVI年变化率＞0.0005 NDVI/a(增加),6%地区NDVI年变化率＜-0.0005 NDVI/a(下降):按照植被覆盖类型,除常绿林、高山草甸年均NDVI呈一定的上升趋势,变化率分别为0.0014 NDVI/a(P0.05=0.001),0.0009 NDVI/a(P0.05=0.001),落叶林、草原、作物、草原化荒漠NDVI没有显著变化(P0.05＞0.05);②年均NDVI与降水、温度相关性分析结果表明,49.00%的地区年均NDVI与年降水量呈正相关,52.33%的地区NDVI与春季降水量正相关,33.69%的地区NDVI与夏季降水量正相关,70.00%的地区年均NDVI与各季气温弱相关,仅17.78%的地区年均NDVI与年均气温正相关;6种植被类型NDVI与降水、气温相关关系为,常绿林、高山草甸年均NDVI与年均气温分别低度、显著正相关性,相关系数分别为0.432(P0.05=0.05)、0.557(P0.05=0.009);草原、作物与年降水量分别显著、低度正相关,相关系数分别为0.5111(P0.05=0.018)、0,476(P0.05=0.029);落叶林NDVI与夏、冬季降水量低度正相关,相关系教分别为0.415(P0.05=0.061)、0.461(P0.05=0.035);草原化荒漠NDVI与春季降水量正相关但不显著,相关系数为0.415(P0.05=0.061).
|
| 关 键 词： | 植被指教 气候因子 趋势分析 相关分析 中亚 |
Relationship between NDVI and Precipitation and Temperature in Middle Asia during 1982-2002 |
| |
| SUO Yuxi,WANG Zhengxing,LIU Chuang and YU Bohua.Relationship between NDVI and Precipitation and Temperature in Middle Asia during 1982-2002[J].Resources Science,2009,31(8):1422-1429. |
| |
| Authors: | SUO Yuxi WANG Zhengxing LIU Chuang and YU Bohua |
| |
| Abstract: | The five countries in Middle Asia lie in the center of Eurasia. Most part of this region is arid and semi-arid zone with sparse vegetation cover. The study of the vegetation dynamics and environmental change in this region is important to the research of environment and climate in China. This paper explored the vegetation dynamics and its relationship with major climatic factors in middle Asia by using AVHRR-NDVI dataset at 8km spatial resolution and CRU climate data set at 0.5° spatial resolution between 1982 and 2002. These two datasets were unified to the same spatial resolution of 8km and Alberta geographic projection. The trend analysis showed that 53 percent of the land cover was relatively stable, with a very small NDVI change of ±0.005 NDVI per year. These regions, especially the two large deserts, were mainly in the center of Middle Asia. Forty percent of the land had a NDVI up-trend of more than 0.0005 NDVI per year, which was mainly in the north and south of Middle Asia, while only 6 percent of the land had a NDVI down-trend of less than 0.0005 NDVI per year. The analysis on land cover types indicated that evergreen forest and alpine grass (steppe) were among the best up-trend group with NDVI gains more than 0.0014 and 0.0009 per year, while the p values are 0.001 and 0.001 respectively. There were no obvious changes in deciduous forest, grass, crop and steppified desert. To investigate the possible driving forces, correlation analysis was conducted between AVHRR-NDVI and major climatic factors, which are precipitation and temperature. In 49 percent of the area, especially in the forest steppe in north Middle Asia, annual average AVHRR-NDVI was closely related to the annual precipitation, especially that in spring and summer. Only 17.78 percent of the area is related to the annual average temperature with a validation coefficient of more than 0.05. Annually speaking, the positive correlation coefficient of evergreen forest, alpine grass with the annual average temperature is relatively low, with the correlation coefficients of 0.432 and 0.557 as well as p value of 0.052 and 0.009 respectively. The positive correlation coefficient of crop and grass with annual precipitation are comparatively low with R values of 0.511and 0.476 as well as p values of 0.018 and 0.029 respectively. The R value between NDVI and precipitation for deciduous forest was 0.415 in summer and 0.461 in winter, while the p value was 0.01 in summer and 0.461 in winter. The positive correlation coefficient of re-vegetated desert cover with precipitation in spring is relatively lower with the R value of 0.415 and the p value of 0.0061. |
| |
| Keywords: | NDVI Climate factor Trend analysis Correlation analysis Middle Asia |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《资源科学》浏览原始摘要信息 |
| 点击此处可从《资源科学》下载免费的PDF全文 |
|
