hide
Free keywords:
ANTHRO-BGC model
Eddy flux measurements
Sensitivity analysis
Parameter optimization
Uncertainty analysis
BIOME-BGC model
terrestrial ecosystem model
carbon-dioxide exchange
biome-bgc
ribulose-1,5-bisphosphate carboxylase
stomatal conductance
parameter-estimation
nitrogen nutrition
leaf-area
wheat
phenology
Abstract:
Climate change and agricultural development are interrelated processes, both of which take place on the global scale. An accurate representation of crop phenology and physiology in ecosystem models is important in order to investigate the interactive processes between atmosphere and biosphere. A new phenological model for crops has been recently incorporated into the ecosystem model BIOME-BGC. Fruit compartments as well as a module for harvest were added to the original model. Here we explore whether eddy flux measurements of agricultural ecosystems are helpful to identify the spatially generalized ecophysiological parameters of the ANTHRO-BGC model (updated BIOME-BGC model) for croplands. The maximum, minimum, and reference values of the ANTHRO-BGC parameters were derived from a literature review. The sensitive parameters of the model were detected by using global sensitivity analysis. The spatially generalized values of ecophysiological parameters for wheat, barley, and rape within a typical temperature and precipitation range in western and central Europe were identified with the help of optimization algorithms and eddy covariance measurements of carbon and water fluxes of agricultural ecosystems as observational constraints. Gross primary productivity (GPP) provided the best constraint on the model parameters for all the three crop types. The validity range of the parameter values defined by the literature review was proved to be reliable. The predictive ability of the optimized model has been improved for GPP and net ecosystem exchange (NEE), but not for evapotranspiration (ET). The performance of the ANTHRO-BGC model was comparable to the performances of other crop models such as DNDC, CERES-EGC, ORCHIDEE-STICS, and SPA model, tested on multiple sites. The optimized ANTHRO-BGC model was relatively stable in response to the variation of the parameters on multiple sites and years. (C) 2011 Elsevier B.V. All rights reserved.
