中国区域典型植被长时间序列光谱数据集

长时间序列植被反射光谱测量在中国的典型农业试验点（包括长春、张掖和海北藏族自治州）持续开展。2016年在长春测量了包括玉米、水稻、小麦、土豆、大豆和绿豆等几种典型植被类型光谱，采样天数从10天到16天不等，分布在整个生长季节。2013年在张掖测量了植被类型包括油菜、油菜、玉米和洋葱等几种植被光谱，采样天数在9到12天，试验期间，每个植被类型每个月至少有两次反射光谱测量。海北藏族自治州的测量观测于2013年，观测了5月至9月共13天的草原反射光谱数据。每个特定目标的光谱为重复观测的平均值。为了收集平滑连续的光谱曲线，应用最小二乘多项式拟合算法去除了大气水汽汽吸收波段的异常波动，并使用分段线性插值函数将光谱曲线插值到1nm光谱分辨率。该数据集有助于植物生长和植被物候的建模，以及生理化参数的反演，另外，还可以作为真实性检验的相对“真值“。

Long-term sequential vegetation reflectance spectral measurements have been continuously conducted at typical agricultural experimental sites in China, including Changchun, Zhangye and Haibei Tibetan Autonomous Prefecture. In 2016, spectra of several typical vegetation types including maize, rice, wheat, potato, soybean and mung bean were measured in Changchun, with sampling intervals ranging from 10 to 16 days across the entire growing season. In 2013, spectra of multiple vegetation types including rapeseed, rapeseed, maize and onion were measured in Zhangye, with sampling intervals of 9 to 12 days. During the experiment, at least two reflectance spectral measurements were conducted per month for each vegetation type. Measurements in Haibei Tibetan Autonomous Prefecture were carried out in 2013, collecting grassland reflectance spectral data over 13 days from May to September. The spectrum of each specific target is the average of repeated observations. To obtain smooth and continuous spectral curves, the least-squares polynomial fitting algorithm was applied to remove abnormal fluctuations in atmospheric water vapor absorption bands, and the piecewise linear interpolation function was used to interpolate the spectral curves to a spectral resolution of 1 nm. This dataset facilitates the modeling of plant growth and vegetation phenology, as well as the inversion of physiological and biochemical parameters. Additionally, it can serve as a relative "true value" for authenticity validation.

搜集汇总

背景与挑战

背景概述

该数据集包含中国长春、张掖和海北藏族自治州等典型农业试验点的长时间序列植被反射光谱测量数据，涵盖多种植被类型，时间跨度为2013年至2016年。数据经过最小二乘多项式拟合和分段线性插值处理，适用于植物生长建模和植被物候研究。