利用红边参数反演植被特征是定量遥感研究的热点问题之一,红边参数中的红边位置与植被参数显著相关,是监测植被胁迫的一个非常敏感的指标。采 用最大一阶导数法、拉格朗日内插法、线性外推法、四点内插法、倒高斯模型法、多项式拟合法六种方法分别计算出不同浓度CO2 泄漏胁迫下大豆 叶片的红边位置,分析并比较不同算法提取的红边位置变化特征,以确定监测CO2 泄漏胁迫下大豆叶片的最佳红边位置的算法。结果表明: 提取的红边位置均与大豆叶片叶绿素含量呈极显著的线性相关。其中,最大一阶导数法和拉格朗日内插法相关程度最高,且最大一阶导数法计算 简单、稳定性较好,因此可以利用最大一阶导数法提取的红边位置反映大豆叶片叶绿素含量变化,进而监测CO2 地质储存泄漏问题。

Red edge parameters are widely used to invert vegetation parameters in quantitative remote sense. The red edge position, as a very sensitive indicator for monitoring vegetation stress, is strongly correlated with vegetation biochemical components. In order to obtain the best red edge position algorithm, six red edge position extraction methods, which are red edge position maximum first derivative method, Lagrange method, line extrapolate method, four-point interpolation method, Gaussian method and polynomial fitting method, were compared for soybean leaf under higher CO2 stress. The results show that the different algorithms of red edge position are significantly linear correlation with chlorophyll content of soybean leaf. However, largest first derivative method and Lagrange method are the optimal extract methods to calculate red edge position for soybean leaf under CO2 stress. Moreover, the maximum first derivative method is more simple and stable. The results imply that red edge position changes can reflect plant chlorophyll content and can be used to monitor CO2 leakage during CCS project using aboveground plant remote sensing data.