The Nonlinear Terrestrial Interactions Investigated by a Multi-layer Higher order closure Ecosystem Model

The terrestrial carbon sink is not only a major carbon removal mechanism accounting for approximately one third of the annual global carbon sink in the atmosphere, but also it is the most important driver shaping the canopy structure for the naturally vegetated land cover. Several studies have found that the changes in canopy structure can regulate ecosystem responses to micro environmental conditions, and the inclusion of this feedback mechanism is imperative to reducing the uncertainties for future climate projections. In this study, we improved a multi-layer higher order closure ecosystem model by coupling the biogeophysical features in the Advanced Canopy-Atmosphere-Soil Algorithm (UCD-ACASA) and the biogeochemical processes in the version 4.5 of the Community Land Model (CLM4.5). With this approach, we were able to improve land surface processes commonly used in current ecosystem models with more realistic turbulent transport scheme and canopy structure representation.

A series of simulations were conducted across diverse ecosystem types at the site level scale, and the simulation results were extensively examined with Leaf Area Index remotely sensed by the Moderate Resolution Imaging Spectroradiometer (MODIS) and AmeriFlux eddy covariance measurements. The effects of carbon-nitrogen cycles and dynamic vegetation development on ecosystem simulations were also investigated. The results show that the coupled model is able to reasonably represent ecosystem responses as accurate as the simulations driven by a carefully quality controlled canopy structure dataset, though some biases are detected. Our investigation implies the need to include realistic and detailed terrestrial ecosystem processes in future earth system simulations to reduce the uncertainties from inappropriate ecosystem representation.