The formulation of pasture seed mixture from a diverse pool of six forage species : A thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy at Lincoln University

Abstract

Optimising pasture yield and quality is needed to meet the global food demand. Pasture sward ecosystems that create beneficial diversity and productivity effects will contribute to this. This research used a multi-species pasture mixture experiment to identify optimal seed mixture species combinations under irrigated conditions. This reseach underpins the machanism of linking species mixed pasture properties to the beneficial diversity attributed to community responses. A large-scale diversity experiment used 69 mixtures from six pasture species to investigate the impact of different functional gropus. A simple mixture of combinations of 2 to 3 species from a grass and legume functional group were the key components. On average, mixture communities produced 16% higher biomass yield and contained 64% lower weed biomass than the average performance of the monoculture swards evaluated from 2018 to 2021. Pasture sward responses differed depending on the associated component species, and over time. The model identified several binary and ternary mixture combinations for improved biomass production, weed suppression, and maximising quality; crude protein (CP) and metablizable energy (ME). The highest contributions for herbage biomass were from binary and tertiary mixes of perennial ryegrass (PR), cocksfoot (C), white clover (WC) and red clover (RC), and plantain namely; PR*WC, PR*RC, C*RC, P*WC*RC, PR*WC*RC, PR*C*P and PR*RC*SC. Among those mixture communities, PR*WC and PR*RC maintained the greatest productivity over three years, giving an annual average herbage biomass yield of 13.6 and 15.2 t/ha, when no nitrogen fertilizer was applied. The mixture components in the ternary mixture (PR*WC*RC) produced an annual biomass yield of 15 t/ha. The mixture effect on weed suppression was strong and existed in several mixture communities even though increasing species richness continuously reduced unsown species biomass. Diversity contribution to the nutritional composition of herbage material was not beneficial, but the species' relative abundance in the mixture improved the herbage quality. The optimal mixture combination was PR*RC at the ratio of 50:50 sown seeds equivalent to 12.6 : 13.4 kg/ha (26 kg/ha) viable seed rate. This gave a mean herbage biomass yield per regrowth cycle of 1.98 t/ha (16 t/ha/yr), with a mean weed biomass of 0.13 t/ha/yr, 10.8 MJ ME/kg DM and 20.7% protein content. Further, the ternary mixture PR*WC*RC at the proportion of 45:11:44 equivalent of 11.3 : 0.8 : 11.8 kg/ha (24 kg/ha) had a mean yield per regrowth cycle of 1.97 t/ha (16 t/ha) DM, 0.8 t/ha weed yield, 10.8 MJ ME/kg DM and 20.8% protein. This ternary mixture combination at the ratio of 50:15:35 was equivalent to 12.6 : 1 : 9.4 kg/ha (23 kg/ha) and produced a mean yield of 1.95 t/ha (16 t DM/ha), 0.48 t DM/ha weed biomass, 10.1 MJ ME/ha and CP 20.2%. These higher productive mixtures were from legume and non-legume functional groups, which improved the pasture herbage biomass yield and quality compared with their monocultures. The sown species proportion of swards differed over time based on the functional species composition of the mixtures. Among the monoculture swards, grasses maintained a higher sown species proportion than legumes or plantain. On average, mixture communities suppressed the weeds, and the suppressive effect was increased with the increasing number of species in the mixture. PR*RC binary mixture maintained a higher sown botanical composition proportion over the 3 years than PR*WC. Modelled Relative Growth Rate Differences (RGRD), with the sown proportion, revealed that the equal proportion of the component species in the identified ternary mixture (PR:WC:RC) balanced the competitive growth and maintained the sown proportion until the 3rd year after establishment. This deviated from the optimal seed proportion that maximised the sward yield responses. Further, this experiment identified that the reason the sward yields differed was investigated in the third year and showed this was due to the quantity of fraction canopy light interception. Overall, species mixed pasture swards intercepted a higher fraction of canopy light (PAR). Average pre-grazing fraction light interception value of swards (2020/21) across the monocultures and two species mixture commmunities was modelled to the initial sown species component proportion. This quantified the diversity contribution of addtitonal fraction of canopy light interception. This was higher for PR*WC (0.47) than for PR*RC (0.10). The higher fraction of light interception contributed to the diversity attributed to the higher herbage biomass in PR*RC than PR*WC. The highest pre-grazing fraction of light interception and intercepted PAR energy was in summer/spring when temperature and available light levels were highest. However, the positive contribution from RC over WC appeared in autumn when the fraction of light interception from white clover may have been compromised by its shallower roots. In the third year, radiation use efficiency (RUE) of swards differed based on the functional species in the mixture community. Grasses had a lower RUE than clovers. There was a positive relationship between RUE and temperature for monocultures and binary mixtures, especially PR*WC and PR*RC. This research suggests farmers in irrigated conditions should sow a mixed pasture of 2-3 species that comprise a grass and a clover. The PR:RC mixture (50:50 = 12.6 : 13.4 Kg/ha) and PR*WC*RC in both proportion settings (45:11:44 or 50:15:35 = 11.3 : 0.8 : 11.8 kg/ha or 12.6 : 1.0 : 9.4 kg/ha) improved the pasture herbage biomass yield and quality over time compared with their monocultures. These mixtures provided diversity response that persisted and optimised biomass yield with minimal weed content and maintained the botanical composition which inturn optimised herbage quality.