植物物候遥感监测精度影响因素研究综述

Review of influencing factors of accuracy of plant phenology monitoring based on remote sensing data

基于植物物候的遥感监测对于研究植被对气候变化的响应具有重要的科学价值。本文在阐述植物物候遥感监测原理及其通用技术流程的基础上,分别从植被类型及其所处的地理条件、遥感数据源及其预处理、植物物候遥感识别方法和植物物候遥感监测结果评价4个方面分析了影响植物物候遥感监测精度的因素,并针对当前研究中存在的不足,探讨了提高植物物候遥感监测精度的可行性途径,即建立高分辨率的近地面遥感定点观测及数据共享网络,发展普适性更强的卫星遥感时序数据去噪及植被指数曲线重建方法,寻求稳定性更高的植物物候期遥感识别方法,探索综合运用地面观测、遥感监测与模型模拟实现物候观测空间尺度拓展的可能性。

Monitoring plant phenology with remote sensing data has important scientific value for studying the response of vegetation to climate change. A comprehensive analysis on the influencing factors of accuracy of plant phenology estimation based on principles and general technical processes of remote sensing application in vegetation monitoring was carried out by taking into account the following four aspects： the specific vegetation type and its geographical conditions; remote sensing data and pre- processing; techniques used to identify plant phenometrics; and evaluation of satellite- derived plant phenometrics. Potential methods for improving the accuracy of plant phenology monitoring are thoroughly discussed. These include： building high-resolution nearsurface sensor- derived phenology observation and sharing network; developing universally applicable methods for noise removal of satellite remote sensing time- series data and reconstruction of vegetation index curves;searching more stable methods to estimate plant phenology; and exploring the possibility of synthesizing groundbased observation, remote sensing monitoring, and model simulation to achieve the spatial scaling- up of phenometrics.