Reconstruction of a Cloud-free Vegetation Index Time Series for the Tibetan Plateau

作者：Wen Jun, Zhongbo Su, Yaoming Ma · 发表于：Mountain Research and Development · 年份：2004 · DOI：10.1659/0276-4741(2004)024[0348:roacvi]2.0.co;2 · 被引用次数：34 · 研究领域：Remote Sensing in Agriculture、Plant Water Relations and Carbon Dynamics、Species Distribution and Climate Change

In this investigation, an advanced Harmonic Analysis of Time Series (HANTS) technique has been introduced in order to remove cloud contamination from the satellite-based Normalized Difference Vegetation Index (NDVI) over the Tibetan Plateau. Due to high elevation, Asian monsoon, and the prevailing local convective weather, the probability of cloud occurrence in the daytime is very high over the Tibetan Plateau. It is therefore very difficult to directly obtain cloud-free vegetation index images using satellite optical remote sensing data over the entire Tibetan Plateau. Applying the HANTS algorithm made it possible to successfully replace the cloud contamination in NDVI images by the HANTS synthesized results, not only in temporal variation but also in regional distribution. The reconstructed NDVI images are cloud-free and more reliable; the HANTS-generated parameters are also potentially useful in understanding vegetation evolution and phenological characteristics over the mountainous regions of the Tibetan Plateau.