Testing interactions between radiation, carbon and water cycles using the LBA data
Yongkang
Xue, University of California, Los Angeles, USA, yxue@geog.ucla.edu
Fernando
Henrique
De Sales, University of California, Los Angeles, USA, fsales@ucla.edu
Jim
Collatz, NASA, jcollatz@biome.gsfc.nasa.gov
Xiwu
Zhan, University of Maryland, Baltimore, xzhan@hsb.gsfc.nasa.gov
The interactions between radiation, water, and carbon are a crucial component in determining terrestrial carbon and water fluxes. In this paper, we will present evidence to demonstrate the close relationship between these processes in an integrated climate system using the Simplified Simple Biosphere Model. We tested this enhanced SSiB using observational data from LBA and Boreal sites. The initial results indicated that the model in general produced a higher than normal rate of photosynthesis which led to an overly large transpiration. We examined model performance and found that this was mainly caused by failing to recognize the effect of diffuse radiation and by not considering the sunlit and shaded leaf areas. Furthermore, only direct radiation effect was included in the scaling equation, which was adapted from SiB2.
Diffuse radiation, which arises from atmospheric scattering and from scattering within the canopy, has been shown to have a crucial role in the photosynthetic process. Therefore, we developed a new physically and biologically based parameterization for shading and scaling to more realistically simulate the land/atmosphere interaction processes. The scaling method considers the effects of both direct and diffuse radiations.
We have tested the new method using the observational data from the LBA experiment ( 2000). This new method substantially improved the simulations of daily mean carbon and water fluxes and their diurnal variations, especially in the tropical area. For example, the root-mean-square (RMS) error for the latent heat flux and the carbon flux are reduced from 15.1 w m-2 and 2.9 mmol m-2 s-1, to 45.4 w m-2 and 12.7 m mol m-2 s-1, respectively. Experiments are also conducted to comprehensively test the effects of soil moisture and clouds in estimating the variability of carbon and water fluxes.