Residual correlation and ensemble modelling to improve crop and grassland models

Multi-model ensembles are becoming increasingly accepted for the estimation of agricultural carbon-nitrogen fluxes, productivity and sustainability. There is mounting evidence that with some site-specific observations available model calibration (with vegetation data as a minimum requirement), median outputs assimilated from biogeochemical models (multi-model medians) provide more accurate simulations than individual models. Here, we evaluate potential deficiencies in how represent (in relation to climatic factors) processes underlying complex systems such grassland crop rotations including fallow periods. We do by exploring correlation residuals. restricted distinction between partial full two most relevant stages, i.e. plant only (partial) combination plant, soil physical (full). It introduces evaluates trade-off (1) what practical apply users beneficiaries, (2) constitutes best modelling practice. The lower correlations obtained overall fully calibrated highlight centrality scenario identifying areas structures require further development.