Soil erosion and degradation is a significant problem in macadamia orchards, particularly as shading increases and interrow groundcover diminishes in higher density orchards. A preliminary investigation of low growing groundcovers for orchards on the north coast of New South Wales in the 1980's found Amarillo peanut (Arachis pintoi), Maku lotus (Lotus pedunculatus cv Grasslands Maku) and bahia grass (Paspalum notatum) most suited to krasnozem soil in the Lismore district. Recent work has aimed at determining the effect of groundcover on growth and yield of macadamia compared to normal bare soil management. Growth, in particular of young trees and nut yield and quality of older macadamia appear to be reduced with groundcover.
The macadamia industry, as with other orchard crops in subtropical eastern Australia has traditionally relied on volunteer naturalised grasses such as kikuyu (Pennisetum clandestinum), carpet grass (Axonopus affinis) and common couch (Cyanodon dactylon) with a mixture of herbs and weeds to provide effective ground cover in the early life of the orchard. However, control of competitive growth has relied on expensive management methods of interrow mowing and intrarow herbicides. In addition, these species are not adapted to low light conditions and tend to decline as the canopy closes over in high density orchards.
Perennial, low-growing groundcover species adapted to low light conditions which provide cover for the entire orchard floor are effective for erosion control, which is a major problem along tree rows maintained bare with herbicides. In addition, groundcovers reduce overland movement of fertilisers in major rainfall events, promote increased water infiltration, help to reduce effects of compaction, suppress weed growth and reduce the need for herbicides, increase organic matter which breaks down to humus and increases soil cation exchange capacity and improves soil structure. Legume species can supply some nitrogen through nitrogen fixation in root nodules.
Groundcover selection and evaluation work was conducted at Alstonville as a minor project between 1982-89 using seed from the pasture species introduction program at Grafton Agricultural Research Station. The main aim was to select a general purpose permanent groundcover to suit all orchard types on the North Coast.
The specifications for an ideal groundcover for the orchard floor were: a permanent perennial or regenerating annual species, preferably a legume, maximum height of 20 cm, low light adaptation, non-climbing, no cultivation requirement for establishment, no irrigation requirement in spring, minimal requirement for mowing, and no harbourage of pests and diseases, especially nematodes.
52 accessions and 13 cultivars involving measurement of plant establishment and dry matter production, some assessment of sward height, groundcover and persistence were conducted over 7 years to determine suitable persistent low-growing grasses and legumes.
Aeschynomene americana, Aeschynomene falcata, Aeschynomene villosa, Arachis monticola, Arachis pintoi, Arachis pusilla, Arachis repens, 6 unnamed Arachis sp. accessions, Cassia mimosoides, 3 Chamaecrista rotundifolia cvs., Crotalaria christantiflora, Crotalaria sagittalis, 2 Desmodium canum accessions, 3 Desmodium scorpiurus accessions, 4 unnamed Desmodium sp accessions, Glycine microphylla, Glycine tabacina, Indigofera hirsuta, Indigofera spicata, 2 cultivars Kummerowia striata, Lablab purpureus, Lotononis bainesii, 2 cultivars Lotus pedunculatus, 3 accessions Macroptilium fraternum, 3 accessions Macroptilium lathyroides, 2 cultivars Paspalum notatum, Stylosanthes guianensis, Trifolium repens cv Haifa, Trifolium semipilosum cv Safari, 3 cultivars Vigna luteola, Vigna parkeri cv Shaw, 2 accessions Zornia sp.
A number of species had the desired growth form and ground cover but were not persistent beyond 2 years with grass and weed competition. The most promising low-growing perennial species included the legumes Amarillo peanut (Arachis pintoi), Maku lotus (Lotus pedunculatus cv Grasslands Maku)) and Desmodium scorpiurus; and Bahia grass (Paspalum notatum cv competidor). Shaw creeping vigna (Vigna parkeri cv Shaw) was slow to establish, but formed a significant sward after 3 years. Desmodium scorpiurus had the desired growth habit but had the disadvantage of sticky seed pods, typical of Desmodiums generally. Amarillo peanut was the most successful legume species with respect to suitability as an orchard groundcover. It is persistent, able to compete with weeds and grasses, has low light adaptation, drought and nematode tolerance, is non-climbing and amenable to mowing. Bahia grass has the advantage of being a less aggressive spreader, requires less mowing than kikuyu and is more adapted to low light conditions.
NSW Agriculture sowed Amarillo peanut in a macadamia orchard at Dorroughby in 1987. Maku lotus was also sown, but was less persistent, sensitive to mowing height, and less suited to macadamia harvest. Amarillo peanut established well in the interrow of mature bearing trees and in 2 years spread under the tree canopy following high tree skirting. It was very effective for erosion control, especially when combined with low-profile mounding, and the new mechanical harvesters recovered nuts easily from the mown sward. However, the species required regular mowing in late summer/autumn to control growth for nut harvest.
Legumes tend not to persist alone in natural systems because their nitrogen fixation encourages grass associations.
In 1988 a replicated small plot legume/grass combination experiment consisting of Amarillo peanut and Shaw creeping Vigna was either sown alone or in combination with Noonan kikuyu, Competidor bahia grass or bahia grass CPI 23944. After 3 years as a mown sward, Competidor bahia grass and CPI 23944 bahia grass formed fairly equal components of grass and legume with Amarillo peanut, whilst the peanut/kikuyu combination had an 80% grass component. Shaw Vigna did not establish and survive with or without grass.
The experiment indicated that Amarillo peanut combined well with grasses and was able to persist in a mixed mown sward.
The main aim of current work is to determine the effect of Maku lotus and Amarillo pinto peanut on growth, nut yield and nut quality of macadamia compared to normal bare soil management along the tree row. A particular emphasis was placed on the measurement of the macadamia and groundcover root systems to determine the effect of groundcover competition on macadamia fibrous root length. The effect of groundcover on soil physical and chemical status over 4 years was also assessed. A glasshouse trial and field trials at three sites was carried out as part of the study.
The aim was to compare the effect of mown and unmown Amarillo peanut with bare soil under 'wet', 'medium dry' and 'dry' water regimes on macadamia growth and water status in large pots over 8-10 months. Banana was included in the experiment as a more water-sensitive species for comparison with macadamia, especially as field work at Alstonville had shown that unmown Amarillo peanut affected banana growth, plant cycling and fruit yield (Johns, 1994).
Plant growth, leaf area, plant water loss as measured by stomatal conductance, total plant dry weight and root length at harvest was measured.
Stomatal conductance data is currently being analysed and is not presented, and further analysis of other data is required before conclusive results can be presented.
Tentative conclusions from the experiment are:
In summary, banana was more sensitive to soil moisture conditions than macadamia and shows more effects under very dry conditions. Unmown groundcover affect banana growth, even under moist soil conditions, whilst macadamia was only affected when conditions were very dry. Mown groundcover had less effect than unmown groundcover on both banana and macadamia in glasshouse conditions.
Field trials were established at two sites in 1991. The Clunes site consisted of 2 year old trees and 5 year old trees, whilst trees at the Dorroughby site were 11 years old. Trial sites consisted of mown groundcover interrow and replicated treatments of mown groundcover (Dorroughby only), unmown groundcover, and bare soil along the tree row for the width of the canopy.
The following parameters were measured annually, unless otherwise indicated: tree trunk girth, seasonal lateral growth and % new leaf growth, root growth (as measured by soil cores to 20cm, and 1-2 monthly observations of periscope tubes embedded in soil under trees to a maximum 57 cm depth), tree water status from pre-dawn xylem water potential readings (mostly in spring/early summer); nut yield (and quality in 1994); soil strength as measured by soil penetrometer resistance, and changes in soil chemical status as indicated by pH, total nutrients and organic matter.
A third trial site using young macadamia trees (planted 1989) at Alstonville was used to determine the effect of planting macadamia into established Amarillo peanut as a mown interrow sward and unmown sward along the tree row, on trunk girth (1993-5) and water use in spring (1993-4) compared to bare soil. Trunk girth information at this site and from young non-bearing trees in Maku lotus at Clunes is presented in one table, although the comparison is not valid statistically. Statistical analysis of data has not yet been done.
Effects of groundcover on macadamia trunk girth, nut yield and quality at Clunes and Dorroughby are shown in Tables 1 and 2, and water status in spring and lateral growth at Clunes are shown in figures 1 and 2. Soil penetrometer resistance for 1995 is shown in figure 3. Data on tree new leaf growth, root growth, and other soil physical and chemical parameters have not yet been collated. Table 0Z1. Trunk Girth (20 cm above ground).
| Control | |||||
| Unmown Lotus | |||||
| Control | |||||
| Unmown peanut | |||||
| Mown peanut | |||||
Table 2(a) Yield (kg NIS 10%)
| Control | ||||
| Unmown Lotus | ||||
| Control | ||||
| Unmown peanut | ||||
| Mown peanut | ||||
Table 2 (b) Quality
| KR (%) | AGK (%) | |
| Control | 30.5 | 94 |
| Unmown lotus | 30.8 | 94 |
| KR (%) | AGK (%) | |
| Control | 38.5 | 91.9 |
| Unmown peanut | 37 | 68.7 |
| Mown peanut | 37.4 | 84 |
Figure 1. Leaf water potential of 6-8 year old trees in bare soil versus groundcover at Clunes October 1992 to November 1994.
Figure 2. Mean lateral length of 6-8 year old trees in bare soil versus groundcover at Clunes 1992-4.
Figure 3. Mean soil penetrometer resistance to 300 mm depth
in bare soil versus groundcover under 8 year old trees at Clunes
(3 probes = mean of 3 probes at canopy edge, mid-probe = 1 probe
in line with tree trunks).
Mean trunk girth for non-bearing trees at Clunes and Alstonville in 1994 (table 3) and pre-dawn leaf water potential and lateral growth at Clunes (1992-4) (figs 4 and 5) for groundcover versus bare soil treatment are indicated below. Table 3 Trunk girth (cm) for young non-bearing trees
| Clunes | |||
| Alstonville |
Figure 4. Leaf water potential of 2-4 year old trees in bare soil versus groundcover at Clunes.
Figure 5. Mean lateral length of 2-4 year old trees in bare soil versus groundcover at Clunes.
Field trial summary
Firth, D.J. and Wilson, G.P.M. (1995). Preliminary evaluation of species for use as permanent groundcover in orchards on the north coast of New South Wales. Tropical Grasslands 29, 18-27.
Johns, G.G. (1994). Effect of Arachis pintoi groundcover on performance of bananas in northern New South Wales. Australian Journal of Experimental Agriculture 34, 1197-1204.