Lucerne in crop rotations on the Riverine Plains 2. Biomass and grain yields, water use efficiency, soil nitrogen, and profitability

In a field experiment in north-eastern Victoria (average annual rainfall 59 8 mm), the impact of 2-4 years of lucerne growth on the following 3-4 crops was assessed. Controls of continuous lucerne, annual pasture, and continuo us crop were compared with 5 lucerne-crop rotations. Above-ground biomass a nd water use efficiency of lucerne, annual pasture, and crops were assessed , as were the soil N status, grain yields, and profitability of crops after lucerne. Lucerne grew more slowly over the autumn-spring growing season (2 0 kg DM/ha.day) than did annual pastures and crops (41 and 58 kg DM/ha.day, respectively), while over the spring-autumn period, it grew at a mean 26 k g DM/ha.day. The summer growth rates of lucerne were, however, highly varia ble (1-52 kg DM/ha.day). Despite large changes in temperature and water ava ilability over the year, the biomass water use efficiency (WUEB) of lucerne was similar over the winter and summer growth seasons (16 and 10 kg DM/ha. mm, respectively) and averaged 13 kg DM/ha.mm for the whole year. In contr ast, the WUEB of wheat, canola, and annual pasture over their respective gr owth seasons averaged 36, 38, and 26 kg DM/ha.mm. When calculated over a wh ole year, however, they were much closer to lucerne at 23, 14, and 17 kg DM /ha.mm, respectively. Autumn removal of lucerne left soils initially low in mineral N (mean 82 kg N/ha.m depth in April) for the establishment of the first crop, but this w as not reflected in the subsequent N contents of crop biomass and grain. Au tumn mineral N concentrations peaked 1-2 years after lucerne removal (mean 141 kg N/ha.m depth). Yields of first crops after lucerne were strongly dep endent on growing season rainfall. When sowing commenced in a wet year, the y were similar to, or greater than, the control, but when sown in a dry yea r, were substantially lower. When sowing commenced in a wet year, lucerne s upplied additional N for a minimum of 2 crops. At least 3 crops were suppli ed with lucerne N when cropping commenced in a dry year. The inclusion of 2 -3 years of lucerne into a continuous cropping sequence only decreased annu al profitability by $AU40/ha. This work shows that short phases of lucerne (minimum of 3 years) followed by 3-4 crops can provide economically viable options for farmers and produce better hydrological outcomes than current a nnual-plant based cropping systems.