Key variables for simulating leaf area and N status: Biomass based relations versus phenology driven approaches
Résumé
Phenological development can be seen as the biological clock of crops and crop model developers used this timeline for describing various processes associated with leaf area dynamics. The critical shoot nitrogen concentration (cNcrit; a reference to quantify crop nitrogen status), leaf-stem partitioning and specific leaf area (SLA) are often simulated as a function of developmental stages or thermal time driven functions. However, the impact of environmental variability might be better incorporated by considering process formulations related to biomass. Using published and novel data, this paper compares the simulation of processes associated with leaf area dynamics of winter wheat and maize using (a) BBCH (decimal code similar to the scale of Zadoks), xsatge or thermal time scale and (b) alternative formulations based on shoot dry mass (DM) or leaf area index (LAI). For both cNcrit and leaf-stem partitioning, the model based on DM was superior to the developmental approaches, as reflected in lower RMSE. Owing to the allometric link between LAI and DM, leaf-stem partitioning and SLA are not independent. Consequently, LAI explained canopy SLA better than developmental traits. Developmental stages are necessary to determine certain events like the switch from vegetative to reproductive development, but our results indicate that phenology cannot capture the allometric relationship between LAI and DM.