Managing crop residues and nitrogen fertiliser to improve wheat yield potential in water-limited environments

Nitrogen (N) supply to rain-fed crops is becoming increasingly challenging due to the decline in organic N reserves. In low-rainfall wheat cropping systems, low crop N uptake has been linked to asynchrony in soil N supply through mineralisation. This is especially true on sandy soils of south eastern Australia which have a low N supply capacity and are considered highly ‘risky’ in a management context. When the N released from soil and residues is insufficient, and/or the timing of biological supply is not well matched with crop demand, management of N inputs to the soil (i.e. legume residue addition and/or fertiliser N) is essential to achieve yield potential and to return a neutral soil N balance for environmental sustainability. The general aim of this thesis was to improve our understanding of the seasonal pattern of the soil N supply capacity via mineralisation for increased wheat N uptake and grain yield, by combining N inputs from different crop residues (removed, wheat or lupin) and fertiliser N inputs (nil, or low, or high N) in a low-rainfall sandy soil environment. Field experiments were conducted over 2 years (2015-2016) at low-rainfall Kandosols based on-farm in the Mallee environment of South Australia. The temporal patterns of the soil profile mineral N and plant available water to 100 cm depth, wheat aerial biomass and N uptake were measured in both years (Chapter 2). In 2016 we also measured the disease incidence as a key environmental variable. There was 35 kg ha⁻¹ more soil mineral N to 100 cm depth following lupin compared with wheat residues at the end of the fallow in both years. In a below average rainfall season (Decile 4), wheat biomass produced on lupin residues was responsive to fertiliser N input with soil profile mineral N depleted by increased crop N uptake early in the season. In an above average rainfall season (Decile 9), a higher soil mineral N supply increased actual and potential grain yield, total biomass, N uptake, harvest index and water use efficiency of wheat, regardless of the source of N (legume N/fertiliser N). These experiments showed that the combination of lupin residues with N fertiliser application increased soil profile mineral N at early growth stages, providing a greater soil N supply at the time of high wheat N demand, and the inclusion of a legume in the rotation is critical for improving the N supply to wheat, with added disease break benefits (Chapter 2). The 2016 field experiment involved the quantification of decomposition rates and N release from wheat and lupin residues over the fallow and the subsequent wheat crop growing season with and without fertiliser N application. It also involved measurements of the temporal patterns of the surface soil mineral N, potentially mineralisable N, microbial biomass N, dissolved organic N and with temperature and rainfall as key environmental variables in all treatments (Chapter 3). Residue decomposition and N release over the fallow and the wheat growing season was measured in the field using litterbags with wheat or lupin residues. Fertiliser N input treatments at wheat crop sowing time and surface soil N pools were measured at key growth stages. A higher potential N supply to wheat following lupin residues at early stages was evidenced through greater decomposition rates and N release via mineralisation than wheat residues, which resulted in increased surface soil N pools. This experiment showed that when lupin residues are combined with fertiliser N application, the N supply capacity to wheat is improved during the growing season measured as mineralised N, dissolved organic N and potentially mineralisable N, relative to wheat residues combined with fertiliser N The last experiment (Chapter 4) was conducted under controlled conditions to directly assess (using ¹⁵N labelled fertiliser) the role of N fertiliser on the supply of N to wheat N through soil mineral and biological pools. This experiment measured the role of the N fertiliser combined with wheat, lupin, or no stubble incorporation. Wheat plants were grown in a glasshouse and sampled at 3 critical wheat growth stages (tillering, first node, booting) to determine wheat and ¹⁵N uptake. Soil samples were collected at sowing, tillering, first node and booting to determine mineral N, microbial biomass N, dissolved organic N, and potentially mineralisable N on subsets of samples. This study indicated that the presence of early N immobilisation (between sowing and tillering) in all the treatments without ¹⁵N fertiliser limited N availability for wheat uptake in the subsequent period (between tillering and first node), when fertiliser N appeared critical to maximise N supply for plant requirements. It was found that up to 38% of the ¹⁵N fertiliser applied at sowing was incorporated into the soil microbial biomass pool. Therefore, the fertiliser N was critical to relieve short-term inherent N limitations for both plant and microbial growth, and to supply the longer-term biological pools (microbial biomass) to support subsequent mineralisation potential. This study also showed that reducing the energy limitation to the microbial pool through inputs of carbon from stubble was critical to ensure fertiliser N supplied sufficient N to satisfy plant demand later in the growing period. This research contributes to a greater knowledge of the main factors affecting soil N dynamics relative to wheat N nutrition and yield, quantifying the N supply from soil and fertiliser and the N accumulation in wheat biomass (roots, shoots and grain) at critical phenological stages in a low rainfall sand. Further research will require measurements of the contribution of different legumes combined with varying fertiliser N rates for a complete assessment of the impacts that could be achieved, and examination of the effect on the main soil N pools driving N supply to wheat N uptake across several seasons and/or in different soil types.Thesis (Ph.D.) -- University of Adelaide, School of Agriculture, Food and Wine, 202