青藏高原连续日光诱导叶绿素荧光数据集（2000-2018）

该数据集是通过MODIS各通道反射率和SIF观测数据建立神经网络模型，从而得到较高时空分辨率的SIF数据，常作为初级生产力的参考。数据来源于Zhang et al. (2018)，具体算法参见文章。源数据范围为全球，本数据集选取了青藏高原区域。本数据将原本的4天时间尺度数据集成至月数据，加工方法为取月最大值，尽可能达到去除噪声的效果。该数据集常被用作评定植被绿度和初级生产力的时间和空间格局，具有实际意义和理论价值。

This dataset trains a neural network model using MODIS channel reflectances and SIF observation data to generate SIF data with high spatiotemporal resolution, which is frequently used as a reference for primary productivity. The original data source is Zhang et al. (2018), and the specific algorithm is detailed in that paper. The source data covers the entire globe, while this dataset only selects the Tibetan Plateau region. This dataset aggregates the original 4-day temporal scale data into monthly data, adopting the processing method of taking the monthly maximum value to minimize noise as much as possible. It is often employed to evaluate the spatiotemporal patterns of vegetation greenness and primary productivity, holding both practical and theoretical significance.

搜集汇总

背景与挑战

背景概述

该数据集是青藏高原区域2000年至2018年的连续日光诱导叶绿素荧光（CSIF）月数据，空间分辨率为1km至10km，通过神经网络模型结合MODIS反射率和SIF观测生成，主要用于评估植被绿度和初级生产力的时空格局，具有较高的实际应用和理论研究价值。