Carbon-Structural Analysis of Global Land Models (C-SALM)
Abstract
Abstract Better understanding of terrestrial carbon cycle is taking an increased scientific attention in the present era of climate change. Representation of the global carbon cycle is increasingly becoming complex in land models which results in large uncertainties in modeled outputs. Therefore, it is urgent to promote methods for quantitative and critical assessment of the models. Here we apply a systematic computational framework for Carbon-Structural Analysis of Global Land Model (C-SALM). The models used in this study are NCAR's Community Land Models (versions CLM3.5, CLM4.0 and CLM4.5) present in Community Earth System Model (CESM), Australian Community Atmosphere Biosphere Land Exchange (CABLE) and Common Land Model (CoLM) of China. The framework applied in this study facilitates the effective model comparison by decomposing a complex land model into traceable components based on fundamental properties of biogeochemical processes implemented in these models. The framework defines ecosystem carbon storage capacity (Xss) as a product of net primary productivity (NPP) and ecosystem residence time (τE). The τE is determined by (i) baseline carbon residence times (τ‧E), (ii) environmental scalars (ξ), and (iii) environmental forcings (Xia et al., 2013). The τ‧E can be further traced by partitioning coefficients (called vector B) and transfer coefficients (called A & C matrices) of NPP. To compare land models, the steady state annual average outputs were computed using 1990 forcing data at 1x1o resolution. The carbon storage capacity of each model was found to be determined differently which are due to differences present in carbon residence time and environmental scalars. The dependency of ξ was assessed based on temperature (ξT) and water (ξW) scalars. This study explains the carbon model pool structure for each model and identifies the A, B and C elements at each carbon pool. The C-SALM study also evaluates models at major plant functional types (PFTs) level and traces major differences in terms of τE, τ‧E and ξ in each model. Climate forcings which control decomposition rates were found to be different at PFT level for each model. The approximation errors raised due to temporal variations of A, B, C and ξ were evaluated at PFT level for each model. Information from C-SALM is helpful in enhancing the understanding of land model performance and reducing uncertainty in model output. Furthermore, this study has a range of implications for future model development and inter-comparison. Reference: Xia J, Luo Y, Wang YP, Hararuk O (2013) Traceable components of terrestrial carbon storage capacity in biogeochemical models. Global Change Biology, 19, 2104-2116.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFM.B11E0402R
- Keywords:
-
- 0428 BIOGEOSCIENCES Carbon cycling;
- 0414 BIOGEOSCIENCES Biogeochemical cycles;
- processes;
- and modeling;
- 0426 BIOGEOSCIENCES Biosphere/atmosphere interactions