Modelling seasonality of dry matter partitioning and root maintenance respiration in lucerne (Medicago sativa L.) crops

In lucerne, a dynamic fraction of total daily assimilates is allocated to the root system and the amount partitioned differs in response to seasonal environmental stimuli. This study proposes a simple framework to quantify this dynamic for inclusion in lucerne simulation models. The framework incorporates recent findings on radiation-use efficiency, radiation interception, and dry matter (DM) partitioning to shoots and roots in response to environmental factors. Results showed that the root mean squared deviation (RMSD) between observed and simulated values was 20% of the mean for shoot biomass and 24% for leaf area index. However, there was systematic bias (73% of RMSD) between observed and simulated root biomass (RMSD=35%). This was caused by the initial assumption of a constant rate for root maintenance respiration (Rm) of 0.015 g/g.day in the model calculations. Sensitivity analysis, with constant Rm values that ranged from 0.005 to 0.030 g/g.day, failed to improve the prediction of root biomass. The modelling exercise suggested the existence of a systematic pattern of change in Rm from 0.035 g/g.day in summer to <0.005 g/g.day in winter. When this seasonality was incorporated, the accuracy of root DM simulations improved (RMSD=14%). The possible existence of a seasonal Rm in lucerne, unveiled by this modelling approach, needs to be validated with an independent dataset.