Prospects for improving productivity and composition in pasture/crop rotations by changing row configuration at sowing

The challenge of maintaining a diverse and robust mixture of pasture species in the rain‚ÄövÑv™fed, semi‚ÄövÑv™arid environments common across south‚ÄövÑv™eastern Australia has long been a topic of interest to farmers and researchers alike. The competition for scarce water and nutrient resources, particularly during summer, is often a key factor driving the decline of desirable species over a relatively short timeframe. Managing interspecific competition by changing the spatial arrangement of different pasture species at sowing could be a practical solution that requires little increase in expense or management complexity for the farmer. However, few studies have investigated the effectiveness of this strategy in improving the productivity of pasture swards, nor examined the broader implications of this approach to farming systems where pastures are grown in phased rotations with crops. This thesis draws on data from a network of field sites comprising two series of experiments. The first, located in the Riverina region of southern New South Wales (NSW), Australia, compared ten pasture swards containing lucerne (alfalfa; Medicago sativa L.), phalaris (Phalaris aquatica L.) and/or subterranean clover (Trifolium subterraneum L.) in various drill row configurations. The second, located in the Central West region of NSW compared legume establishment under cover crops of wheat (Triticum aestivum L.), barley (Hordeum vulgare L.), canola (Brassica napus L.) and lupin (Lupinus angustifolius L.) sown in mixed and alternate drill row configurations with lucerne and subterranean clover. These investigations aimed to evaluate the impact of changing row configuration at sowing on pasture productivity and persistence, and examine the legacy of this practice on soil fertility, the microbiome as well as subsequent crop production. Sowing species into fewer drill rows often led to reduced overall biomass, although results varied from site to site with differences less evident under drier seasonal conditions. Less total biomass was attributable to a reduction in the productivity of the dominant species in the mixture. Herein lies the ultimate challenge when managing competition in a mixture by increasing the spatial distance between species at sowing. On the one hand the spatial separation of species reduced early competition to improve the opportunity for less vigorous species to establish. On the other hand, constraining the dominant species to fewer drill rows limited the productivity of that species, which frequently resulted in reduced overall productivity of the sward. This was because constraining species to fewer drill rows effectively increased the spacing between those drill rows, and the yield of companion species in the intervening rows often could not compensate for the yield reduction of the dominant species. Nowhere was this seen more clearly than in a covercrop scenario where the yield reduction associated with confining the vigorous winter crop to alternate drill rows was not matched by the production of slow‚ÄövÑv™establishing pasture legume seedlings in every other drill row. The legacy of row configuration on perennial‚ÄövÑv™based pasture productivity is enduring. All three pasture species tested were observed to remain largely confined to the original drill for the duration of the 3‚ÄövÑv™year pasture phase, although a greater proportion of the self‚ÄövÑv™regenerating subterranean clover was found on the inter‚ÄövÑv™row area (up to 30%) compared to the perennial species (<5%). This was less evident at the hotter and drier sites where <10% of subterranean clover plants were found in the inter‚ÄövÑv™row area. As a result of remaining confined to the drill row, ongoing reductions in the productivity of dominant species were observed for the life of the pasture phase. For example, at sites where lucerne was the dominant species, row configuration that confined lucerne only to alternate rows had lower ongoing productivity compared to where lucerne was sown in every drill row on account of the increased spacing between lucerne rows in the alternate drill row treatment. However, there was evidence that companion species such as subterranean clover benefitted from spatial separation at sowing. Subterranean clover herbage yield and seed production was consistently greater in treatments where it was sown alone in fewer drill rows compared to where it was sown in every drill row with perennial and covercrop species. In examining the legacy of pasture row configuration, it was clear that regardless of which species were sown, the soil fertility was substantially different immediately under the pasture drill row compared to in the inter‚ÄövÑv™row area. Soil organic carbon (SOC), mineral nitrogen (N), pH and available (Colwell) potassium were all higher on the drill row, plant‚ÄövÑv™available phosphorus was lower and there was a variable response in sulphur with depth. Given the absence of fertiliser application in the drill row at sowing, the spatial differences in fertility at the end of the pasture phase were plant‚ÄövÑv™driven responses, associated with the ongoing concentration of plants on or near the drill row over the life of the pasture. A corresponding difference in soil microbial populations was also observed with a greater abundance and diversity of fungal and bacterial populations on the drill row. This was attributed to the increase in total C concentrations in the surface soil, implicating plant cover as an important driver of improved soil chemical and biological attributes. Ongoing differences in soil chemistry were observed over time following the termination of pastures with herbicide and although some differences between the drill and inter‚ÄövÑv™row areas diminished, levels of SOC remained significantly higher on the pasture drill row (2.34%) compared to between drill rows (2.12%) in the surface 50 mm of soil, 15 months after the pasture was removed, despite a wheat crop being grown in the meantime. Differences in soil fertility at the end of the pasture phase led to an 11% increase in grain yield in wheat growing on the previous pasture drill row compared to plants growing on the interrow area. However, this finding is likely to be of little practical significance, because when averaged over the whole plot area, there was no significant difference in yield due to row configuration at any of the three sites tested. Grain size was smaller and protein concentration was higher in wheat following pastures that had a higher lucerne and/or subterranean clover content due to increased N inputs associated with biological N\\(_2\\) fixation. Taken together, these results suggest that changing drill row configuration does offer potential to increase the subterranean clover content in mixed pasture swards, which has benefits to grain quality in subsequent wheat crops. However, the broadscale use of changing row configuration at sowing is not recommended due to the high probability of reduced overall productivity. Rather, a targeted approach is required where knowledge of the dominance hierarchy of species in the mixture exists. Perverse outcomes on total productivity will likely be minimised where row spacing is reduced as narrower drill rows would result in a lower yield gap in crops and pastures due to a larger number of drill rows being sown to the dominant species for a given area of land. Narrower drill rows will achieve greater plant coverage over the land area in short‚ÄövÑv™term pasture phases, which has been demonstrated to confer benefits to soil chemical and biological properties.