Quantifying the effects of stress on lucerne (Medicago sativa L.) growth and development : A thesis submitted in partial fulfillment of the requirement for the Degree of Doctor of Philosophy at Lincoln University

Abstract

The main aim of this research was to generate data that could be used for further calibration of the APSIM_lucerne model to account for plant stresses. To do this, six experiments were established using lucerne (Medicago sativa L.) at Lincoln University, New Zealand. These experiments investigated the effects of winter grazing management on the spring growth and development of lucerne, the effects of phosphorus on growth and development of lucerne, and the effects of water deficits on the growth and development of lucerne. The trigger for the start of stem extension each spring was shown to be an ~11 h photoperiod, irrespective of the thermal time accumulation since the last defoliation event prior to that point. Incorporating this trigger point and a thermal time based rate of stem extension pre (8.2 oCd/mm) and post (1.17±0.01oCd/mm) an 11 h photoperiod into APSIM_Lucerne improved spring stem height predictions (R_RMSE decreased from 121% to 47.9%) and validated the previously optimised functions. This finding enabled improved winter and spring grazing management guideline for rainfed sheep grazing systems. A late (01/08) winter “clean-up” graze did not affect spring growth so offers flexibility to farmers, but it did reduce root reserves by ~1 t/ha, therefore a longer spell in autumn is required to maintain stand life. In spring, entering Paddock 1 of a six paddock rotation at ~100 mm allows for entry heights in sequential paddocks to be staggered to maintain feed quality. Under a grazed, dryland system lucerne shoot biomass, NDVI, and water use efficiency were unaffected by Olsen P levels of 10 or 20 mg/l over four years when grown in a silt loam soil with a pH of ~6.0. Under an irrigated cut and carry system lucerne maintained yields between 11500 and 19500 kg DM/ha/yr at Olsen P values as low as ~2 mg/l, which suggested that an Olsen P of 2 mg/l in the top 150 mm was sufficient. Results suggest the deep tap root system, and root adaptations allowed the stands to access subsoil P, which was not accounted for in the current soil testing recommendations. Soil and rooting depth should be considered for more efficient P fertiliser recommendations. A pot experiment showed growth responses to P deficiency but no impact on crop development. This suggests that the plants restricted organ size to maintain P concentrations of ~0.16-0.33%, rather than change the number of new organs initiated. Further investigation is required to determine if this was a direct response to inadequate P or if the P reduced N availability which could be expected to produce the same plant response. Mild water stress increased canopy temperatures which accelerated crop development but continued severe stress caused development to cease. This in turn reduced canopy size through a reduction in plant height (33-60%), node appearance (40-61%), leaf area (63-67%), and shoot biomass (30-60%). Functions were developed in APSIM_Lucerne for heightchron, phyllochron, and LAER which showed there were no impacts of water stress until a T/TD of 0.7. This is consistent with values reported in the literature, and with further validation could be implemented into the APSIM_Lucerne model. Overall, this thesis provided experimental data to advance the calibration of APSIM_Lucerne for spring growth and development, P use, and water deficits. However, further validation with independent datasets is required. These findings also have practical implications for grazing management, P fertiliser recommendations, and feed availability through drought.