Accounting for canopy structure improves hyperspectral radiative transfer and sun-induced chlorophyll fluorescence representations in a new generation Earth System model
Accounting for canopy structure improves hyperspectral radiative transfer and sun-induced chlorophyll fluorescence representations in a new generation Earth System model
复制标题
DOI：
10.1016/j.rse.2021.112497
复制
发表时间：
2021-05-13
影响因子：
13.5
通讯作者：
Frankenberg, Christian
中科院分区：
文献类型：
--
作者：
Braghiere, Renato K.;Wang, Yujie;Frankenberg, Christian
三维（3D）植被冠层结构在辐射与地表相互作用的方式中起着重要的作用。在地球系统模型（ESM）中准确地表示这一过程对于全球碳，能量和水循环的建模以及未来气候预测至关重要。尽管考虑三维冠层结构的重要性，但无法在区域和全球范围内表示这种复杂性，阻碍了成功实施ESM。另一种方法是使用隐式聚类指数来解释全球尺度上ESM中植被冠层表示的水平异质性。本文评估了如何建模的高光谱短波辐射分区的陆地生物圈，以及太阳诱导叶绿素荧光（SIF）的影响时，成团指数参数化纳入辐射传输计划的新一代ESM，气候模式联盟（CliMA）。在ESM中精确的高光谱辐射传输表示对于准确地使用卫星数据来对抗、约束和改进陆地模式过程至关重要。新实施的计划相比，蒙特卡洛计算理想化的场景，从辐射传输模型相互比较的项目相互比较的地表参数化（RAMI 4PILPS），为开放的森林冠层和没有雪在地面上。结果表明，在计算高光谱辐射传输时，考虑冠层结构异质性是至关重要的。与不考虑聚集的情况相比，短波辐射的反射率（25%）、吸收率（66%）和透射率（75%）的RMSE降低。计算的SIF与最近发射的NASA轨道碳观测站（OCO）3的卫星遥感数据进行验证，表明在推导SIF时包括垂直和水平的冠层结构可以提高模型预测高达51%的情况相比，没有结块。通过在气候-陆地模型中加入一个聚集指数，可以详细研究冠层结构与SIF、GPP、高光谱辐射传输和冠层尺度观测几何之间的关系。
Three-dimensional (3D) vegetation canopy structure plays an important role in the way radiation interacts with the land surface. Accurately representing this process in Earth System models (ESMs) is crucial for the modeling of the global carbon, energy, and water cycles and hence future climate projections. Despite the importance of accounting for 3D canopy structure, the inability to represent such complexity at regional and global scales has impeded a successful implementation into ESMs. An alternative approach is to use an implicit clumping index to account for the horizontal heterogeneity in vegetation canopy representations in ESMs at global scale. This paper evaluates how modeled hyperspectral shortwave radiation partitioning of the terrestrial biosphere, as well as Sun-Induced Chlorophyll Fluorescence (SIF) are impacted when a clumping index parameterization is incorporated in the radiative transfer scheme of a new generation ESM, the Climate Model Alliance (CliMA). An accurate hyperspectral radiative transfer representation within ESMs is critical for accurately using of satellite data to confront, constrain, and improve land model processes. The newly implemented scheme is compared to Monte Carlo calculations for idealized scenes from the Radiation transfer Model Intercomparison for the Project for Intercomparison of Land-Surface Parameterizations (RAMI4PILPS), for open forest canopies both with and without snow on the ground. Results indicate that it is critical to account for canopy structural heterogeneity when calculating hyperspectral radiation transfer. The RMSE in shortwave radiation is reduced for reflectance (25%), absorptance (66%), and transmittance (75%) compared to the scenario without considering clumping. Calculated SIF is validated against satellite remote sensing data with the recently launched NASA Orbiting Carbon Observatory (OCO) 3, showing that including vertical and horizontal canopy structure when deriving SIF can improve model predictions in up to 51% in comparison to the scenario without clumping. By adding a clumping index into the CliMA-Land model, the relationship between canopy structure and SIF, Gross Primary Productivity (GPP), hyperspectral radiative transfer, and viewing geometry at the canopy scale can be explored in detail.