基于两种多路径参数的GNSS-MR植被指数反演特性对比与分析

Comparison and Analysis of GNSS-MR Vegetation Index Inversion Characteristics Based on Two Multipath Parameters

  • 摘要: 归一化差异植被指数(normalized difference vegetation index, NDVI)是监测植被变化的重要参数之一。随着全球导航卫星系统(global navigation satellite system, GNSS)的不断发展,GNSS多路径反射测量技术被证明具有监测植被变化的潜力。基于GNSS信噪比数据提取的归一化振幅Anorm(normalized amplitude)和基于GNSS多路径数据提取的归一化微波反射指数(normalized microwave reflectance index, NMRI)是两类常用的NDVI反演参数。首先,利用板块边界观测网中P042、P052和P675 3个站点长达12年的GNSS数据进行实验,发现Anorm、NMRI与NDVI之间均存在良好的线性相关关系,相关系数介于0.58~0.86。然后,利用第五代欧洲中期数值预报中心大气再分析数据集及干旱数据,探讨气象干旱与NDVI、Anorm和NMRI之间的对应变化关系,当降水量较少、气候干旱时,对应的NDVI、Anorm和NMRI值均减小。最后,进一步提取了GPS、GLONASS和Galileo不同频点信号对应的NMRI参数,分析其与NDVI之间的相关性。结果表明,当站点周围植被类型为草地和农作物时,三大系统不同频点的NMRI均与NDVI呈现较好的对应关系,多模多频GNSS信号被证明具有植被指数可监测性。

     

    Abstract:
    Objectives The normalized difference vegetation index (NDVI) is one of the important parameters for monitoring vegetation changes. With the continuous development of the global navigation satellite system (GNSS), GNSS multipath reflectometry technology has been proven to have the potential for monitoring vegetation changes. This paper aims to explore the correlations between NDVI and the normalized amplitude (Anorm) and normalized microwave reflectance index (NMRI) obtained from GNSS interferometry reflectometry to assess the application potential of GNSS signals in vegetation monitoring.
    Methods The study used GNSS signal-to-noise ratio data to calculate Anorm and NMRI, analyzing their correlations with NDVI and change trends. In addition, total precipitation data from the European centre for medium-range weather forecasting reanalysis v5 and drought severity data were utilized to explore the relationship between precipitation/drought and Anorm and NMRI. Further analysis was conducted on the correlation between NMRI parameters and NDVI at different frequency points under the GPS, GLONASS, and Galileo systems.
    Results Analysis of experimental data from three plate boundary observatory sites (P042, P052, and P675) reveal strong linear correlations between Anorm and NDVI, as well as between NMRI and NDVI, with correlation coefficients ranging from 0.58 to 0.86. When the surrounding vegetation types were grassland and crops, NMRI at different frequency points across all three GNSS systems demonstrate good correspondence with NDVI. Notably, among the three frequency bands, the L1-band NMRI exhibites the strongest correlation with NDVI, with coefficients ranging from 0.46 to 0.89, which may be attributed to the shortest wavelength of the L1 signal.
    Conclusions The study shows that multi-constellation and multi-frequency GNSS signals can effectively monitor vegetation indices, which validates the application potential of GNSS interferometric reflectometry technology for monitoring vegetation changes and provides important data support for future research.

     

/

返回文章
返回