微波植被指数在云南省植被动态监测中的适用性研究

基于遥感的植被指数是科学监测植被动态变化的最有效方法。然而,在我国西南地形复杂区域,基于植被光谱特征的光学植被指数常常因大气状况及环境条件等的影响而受到很大的限制。利用云南省2013年1月至2018年12月AMSR2 (Advanced Microwave Scanning Radiometer 2,即先进微波扫描辐射计2)双极化亮温数据,计算了云南省2013—2018年多年平均逐月微波植被指数,并选取草地、耕地、落叶阔叶林、常绿阔叶林及常绿针叶林五种典型植被类型区,对比分析了不同植被类型区各微波植被指数的季节变化规律及其与光学植被指数(NDVI)的相关性。结果表明:各微波植被指数的变化幅度均较小,低频和高频微波极化差异指数(MPDI)可以反映云南省各种植被类型的季节变化规律,同时低频MPDI对植被季节变化特征的响应更显著,而低频微波植被指数(MVIA和MVIB)对草地的季节变化响应更敏感。各微波植被指数与NDVI的相关性在低矮植被区更显著,更能反映低矮植被类型随季节变化规律。总体看来,各微波植被指数能够很好地识别不同类型植被的季节变化规律,可作为光学植被指数的有力补充,用于长时序、大范围植被动态监测。

Study on the applicability of Microwave Vegetation Indices in monitoring of vegetation dynamics in Yunnan Province

Background, aim, and scope Vegetation indices derived by remote sensing is the most effective method in monitoring of the long-term vegetation dynamics. However, due to the influence of atmospheric conditions and environmental conditions, the optical vegetation indices based on the spectral characteristics of vegetation often have many limitations, especially in Southwest China with complicated topography. Microwave remote sensing has the advantages of all-day and all-weather observation because of that the long wavelength microwave can penetrate cloud, rain and vegetation canopy. Therefore, the microwave vegetation indices could be theoretically better in monitoring of vegetation dynamics in complex terrain areas with dense vegetation. In this study, taking Yunnan Province as the research area, the applicability of microwave vegetation indices in monitoring vegetation dynamics was been validated in detail. Materials and methods The three microwave vegetation indices, such as microwave polarization difference index (MPDI) and microwave vegetation indices (MVI^A and MVI^B) were calculated using the dual-polarized L3 brightness temperature products of Advanced Microwave Scanning Radiometer 2 (AMSR2). Then the applicability of the microwave vegetation indices was been discussed by comparative analysis of the seasonal variations in different vegetation types and their correlation with Normalized Difference Vegetation Index (NDVI). Results The results show that, the correlation between microwave vegetation indices and NDVI in different vegetation types, as well as their response to seasonal variations have very complicated characteristics. The MVI^A parameter positively correlated with NDVI, while the MPDI and MVI^B negatively correlated with NDVI. Discussion Both low frequency and high frequency MPDI can well reflect the seasonal variations of various vegetation types in Yunnan Province, while the response of low frequency MPDI to the seasonal variations is more significant. Moreover, Low-frequency microwave vegetation indices are more sensitive to seasonal change of grassland. Conclusions The microwave vegetation indices can well identify the seasonal variation of different types of vegetation in Yunnan Province. Recommendations and perspectives The microwave vegetation indices could be used as a powerful supplement to optical vegetation indices for long-term and large-scale monitoring of vegetation dynamics in Southwest China.