Growth of pasture species in the shade in relation to alder silvo-pastoral systems : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy (Ph.D.) in Pastoral Science, Institute of Natural Resources at Massey University, New Zealand
An increased understanding of the competitive interactions between tree species and understorey pastures is required for the development of deciduous tree based silvo-pastoral systems. In particular, the shade tolerance of pasture species likely to be used under trees in New Zealand needs to be determined. This thesis examines the effects of light intensity and quality on the growth of pasture species in a series of glasshouse experiments, and under the shade of alder trees pruned to different heights. The shoot dry weight per plant of all grass and legume species examined showed a linear increase (P<0.0001) with % ambient photosynthetically active radiation (PAR). Highest shoot dry weight was at 73% and lowest was at 14% PAR (heavy shade). Shade also affected the tillering ability of pasture species. Under heavy shade, cocksfoot (Dactylis glomerata L.) produced more tillers per plant than the other grass species examined. Perennial ryegrass (Lolium perenne L.) had the lowest tillering in heavy shade. Under medium shade (43% ambient PAR), tiller number per plant for browntop (Agrostis capillaris L.) and Poa trivialis (Poa trivialis L.) was higher than other species. Lotus (Lotus uliginosus L.) produced a higher (P<0.0001) number of branches under heavy shade than white clover (Trifolium repens L.) and subterranean clover (Trifolium subterraneum L.) Shade affected perennial ryegrass more than cocksfoot selections, especially at the lowest PAR level both in the glasshouse and the field experiment. For example, tillers per plant under tree shade, and also at the low PAR level in the glasshouse for perennial ryegrass were 18 compared with 28-29 (P<0.0001) for Wana cocksfoot and 24-27 for PG 74 cocksfoot. Leaf area per plant for perennial ryegrass was also significantly (P<0.0001) lower than for Wana cocksfoot. Cocksfoot selections were more tolerant of heavy shade than perennial ryegrass, and Wana was the most tolerant of the cocksfoot selections of heavy shade. There were no effects of R:FR ratio (P>0.05) on the shoot dry weight production of the pasture species examined. Similarly, the interaction between PAR x R:FR and species was not significant (P>0.05) for most morphological characteristics when plants were not defoliated. Perennial ryegrass, Wana cocksfoot and Yorkshire fog (Holcus lanatus L.) at low PAR had similar yields, that higher than white clover and lotus, which were similar. However, when plants were defoliated weekly or three-weekly, Wana cocksfoot out produced Nui perennial ryegrass at low PAR/R:FR due to its ability to maintain higher leaf area and higher leaf dry weight, higher SLA, and more tillers per plant. Herbage mass of swards in heavy and medium shade created by pruning alder trees was about 50% and 70%, respectively, of that of light shade (P<0.0001). Herbage mass was highest for cocksfoot either with lotus or white clover (P<0.0001), whereas values for perennial ryegrass and Yorkshire fog were lower and similar. Shade affected perennial ryegrass more than cocksfoot and Yorkshire fog, especially at the lowest PAR level. Cocksfoot in mixture with either white clover or lotus had the highest leaf expansion per tiller, which was in the order cocksfoot> Yorkshire fog>perennial ryegrass. There was no significant difference (P>0.05) between the pasture species in the total number of sheep grazing observations in 2 hours, but more sheep grazed in light shade than in heavy shade (P<0.05). The research highlighted the importance of measuring shade tolerance of pasture species in terms of attributes that determined growth and persistence. As perennial pasture species are regularly defoliated they must be able to vegetatively reproduce in the shade as well as be productive. Shade tolerance of the pasture species examined varied greatly, but their relative shade tolerance was also sensitive to the level of shade. Although, cocksfoot was the most shade tolerant species in heavy shade (PAR level <200 μmoles photons m-2 s-1) it was similar to other species in medium shade (PAR level ≥ 400 μmoles photons m-2 s-1 or more). Light intensity was more important for growth and vegetative reproduction than light quality for pasture species under shade. Likewise, pruning trees was more important for pasture production under tree shade. The morphological attributes related to shade tolerance of New Zealand hill country pastoral plants were identified in this thesis as tiller number per plant, leaf area, specific leaf area (SLA), and leaf appearance interval. For alder silvo-pastoral systems with high tree density and heavy shade (PAR level <200 μmoles photons m-2 s-1) cocksfoot in combination with either lotus or white clover was the most productive pasture, while perennial ryegrass, or browntop, with white clover was as productive as cocksfoot if shade was maintained at a PAR level >200 μmoles photons m-2 s-1). Additionally, cocksfoot and lotus are both tolerant of the low to medium soil fertility and seasonal dry periods likely to be encountered on the hill country where deciduous trees are also used to control soil erosion. Shade had a marked effect on tillering as well as on shoot dry weight, and is the most significant factor determining the understorey pasture production. However, the decrease in pasture production due to shading can be managed by appropriate pruning practices and choice of appropriate pasture species.