喀斯特地区不同植被类型NDVI变化及驱动因素分析——以贵州为例

The Analysis of the Difference Vegetation Variation and Driver Factors on NDVI Change in Karst Region:A Case on Guizhou

以SPOT-VEG NDVI数据为基础结合植被类型、气象和石漠化数据,通过NDVI变化趋势倾斜率及逐像元相关分析,分析不同植被类型NDVI变化趋势及驱动因素。结果表明,(1)2000—2013年贵州省植被NDVI呈增加趋势,其中2000—2007年为快速增加期,变化率为0.25/10 a(r^2=0.923);2008—2013年增速减缓,变化率为0.02/10 a(r^2=0.381)。(2)人工植被NDVI增速最大为0.17/10 a(r^2=0.813),灌丛灌草丛次之,为0.13/10 a(r^2=0.85),乔木类植被(常绿阔叶林、落叶阔叶林、常绿和落叶阔叶混交林、针叶林、针阔混交林)和竹林的NDVI基本保持不变。(3)贵州省气候变化呈不显著冷干趋势,其中降水对植被变化的影响力大于温度,植被NDVI与年降水量和年均温均呈现不显著负相关关系。(4)人工植被与降水和气温的逐像元分析中,显著负相关比重较大,分别达到20%和15%;灌丛灌草丛的显著负相关比重也大于正相关,分别达到16%和17%;乔木类植被则相反,显著正相关比重较大,其中河谷季雨林达到48%。(5)人类活动强度较高的区域,NDVI变化与城市扩展、植树造林及石漠化治理面积有显著正相关性。由此得出,在人类活动强度较大的区域,如城镇周边、生态治理与修复措施的实施区域,植被变化主要受人为作用制约;但当人类活动或干扰较少时,气候变化限制植被的变化趋势。所以,从宏观角度分析植被变化与气候变化的关系时,必须权衡人为作用和气候变化对植被变化的影响。

In order to indicate the trend of vegetation change and the driving factors in Karst region, the slope of normalized difference value index （NDVI） change trend and the correlation analysis of pixel by pixel were used to analyze the NDVI change trend and driving factors of different vegetation based on the SPOT-VEG NDVI data and combined with vegetation type, meteorological data and rocky desertification data. The results showed that： （1） The NDVI of Guizhou Province presented an increase trend from 2000 to 2013, in which it presented a significant increase during 2000 to 2007 and the ratio of change was 0.25/10 a （r^2=0.923）; while the growth slowed from 2008 to 2013 and the ratio of change was 0.02/10 a （r^2=0.381）. （2） The NDVI of artificial vegetation grew fastest with the speed of 0.17/10 a （r^2=0.813）, the secondly was the shrub land and grass with the speed of 0.13/10 a （r^2=0.85）. The NDVI of trees and bamboo forest was almost invariant. （3） The climate change in Guizhou was dry-cool and precipitation influenced more than temperature. There was no significant negative correlation between NDVI and the annual precipitation and the annual mean temperature. （4） By pixel and pixel analyzing the artificial vegetation, precipitation and temperature, the ratio of negative correlation was bigger which reached to 20% and 15% respectively. The negative correlation ratio of shrub land and grass was bigger than positive correlation which reached to 16% and 17% respectively. On contrast, the tree’s positive correlation ratio was bigger in which that of river valley monsoon forest reached to 48%. And （5） in those areas of higher intensity human activities such as urban periphery and rocky desertification management regions, the NDVI change was significant correlated with urban expanding, afforestation and rocky desertification restoration areas. Therefore, in these areas, the vegetation change was mainly attributed to human activities. However when human activities induced less, the climate change restricted the trend of vegetation change. To analyze the relationship between vegetation change and the climate change from the macroscopic angle, the influence weight of human action and climate change must be identified.