Estimating global transpiration from TROPOMI SIF with angular normalization and separation for sunlit and shaded leaves

All three types of SIF-driven T models integrate canopy conductance (gc) with the Penman-Monteith model, differing in how gc is derived: from a SIFobs driven semi-mechanistic equation, a SIFsunlit and SIFshaded driven semi-mechanistic equation, and a SIFsunlit and SIFshaded driven machine learning model. The difference between a simplified SIF-gc equation and a SIF-gc equation is the treatment of some parameters and is shown in https://doi.org/10.1016/j.rse.2024.114586. In this dataset, the temporal resolution is 1 day, and the spatial resolution is 0.2 degree. BL: SIFobs driven semi-mechanistic model TL: SIFsunlit and SIFshaded driven semi-mechanistic model hybrid models: SIFsunlit and SIFshaded driven machine learning model.三类基于日光诱导叶绿素荧光（Solar-Induced Chlorophyll Fluorescence, SIF）驱动的T模型均将冠层导度（canopy conductance, gc）与彭曼-蒙特斯模型（Penman-Monteith model）进行耦合，其核心差异在于冠层导度的推导逻辑：分别采用SIF观测值（SIFobs）驱动的半机理方程、光照SIF（SIFsunlit）与遮荫SIF（SIFshaded）驱动的半机理方程，以及光照SIF与遮荫SIF驱动的机器学习模型。 简化版SIF-gc方程与常规SIF-gc方程的差异在于部分参数的处理方式，具体细节可参见https://doi.org/10.1016/j.rse.2024.114586。 本数据集的时间分辨率为1天，空间分辨率为0.2°。 BL模型：由SIF观测值驱动的半机理模型 TL模型：由光照SIF与遮荫SIF驱动的半机理模型 混合模型：由光照SIF与遮荫SIF驱动的机器学习模型