Site
contents Biodiversity & Restoration paper | You are here: Tussock
Grasslands MIS Site
contents Biodiversity & Restoration paper |
Bill Lowther
AgResearch, Invermay Agricultural Centre, Mosgiel.
IntroductionThe introduction by European settlers of grazing animals to the tussock grasslands in the last century was a major ecological disturbance to which many tussock grassland species are ill adapted. Grazing, combined with regular burning of grasslands to produce new shoots palatable to stock, periodic rabbit infestation and the invasion of exotic weeds (e.g. Hieracium), has reduced both tussock biomass and cover. In addition, species biodiversity in the tussock grassland declined a process that is still continuing (Duncan et al. 2001). In some tussock grasslands environments, accelerated erosion has been attributed to the loss of vegetative cover.
There has been considerable New Zealand research into revegetation of erosion prone tussock country but the emphasis has been placed on using introduced plants often with the application of high rates of fertiliser (Nordmeyer pers. comm.). Eroded tussock grassland soils are very low in plant nutrient, particularly nitrogen, phosphorus and sulphur and the initial emphasis should be placed on introduction of legumes with appropriate fertiliser (Protection Forestry Diversion, Forest Research Institute report, undated). The report concluded that native grasses (un-named) were mainly slow growing. It also suggested that harvesting indigenous seed in bulk would present difficulties. The use of native plants has usually been restricted to areas with particular aesthetic values and has usually involved relatively small-scale plantings (Nordmeyer pers. comm.).
Information on small scale restoration, generally for soil conservation purposes is available (e.g. Pollock 1986), or the establishment of a range of native grasses (e.g. Wills 2002) and is not covered in this review. However, recently, the lack of reliable and cost effective techniques for large-scale restoration of tussock grasslands has become evident. This report reviews information relevant to large-scale restoration of tussock grassland species, with the emphasis on tussock grasses. In some instances, information from establishment techniques for pastoral species has been extrapolated to tussock grassland species.
The extensive natural stands of snow tussock (Chinochloa spp) play an important role in water and soils conservation and in aesthetic values. In areas dominated by snow tussock vegetation, disturbed soils can be revegetated by using local ecotypes. However, the high costs in establishing a reasonable cover is a deterrent. Seed supply is irregular because of sporadic flowering, poor seed set or poor filling in some years and insect predation. Seedlings raised in the nursery require at least 2 years to reach a plantable size of 20 cm or more. Seedlings are best planted in winter or early spring, but if raised in a lowland nursery the seedlings will require some extra hardening off before planting at higher elevations. Tussocks can also be established by dividing vigorously tussocks into clumps of 10+ tillers, planted in a nursery for a year and then transplanting. Larger clumps can be transplanted directly. Plants respond to fertiliser in low fertility conditions.
Fescue tussock(Festuca novae-zelandiae) is common throughout the tussock grasslands where it was frequently the dominant species in the short tussock grasslands of the lowland montane regions. However, it has been declining due to rabbits and Hieracium invasion. It is not a colonizing species and only achieves dominance after long periods of stability. Silver tussock(Poa cita)is representative of higher fertility grassland environments, particularly those in moister areas. Natural reseeding can be encouraged by removal of stock during flowering and natural revegetation can be rapid under suitable conditions. Blue tussock(Poa colensoi)is quite palatable and can be overgrazed by domestic stock and rabbits.Restoration is very dependent on removal of rabbits. These three species can be easily propagated from seed in the nursery although it is recommended that plants are raised for two years in the nursery before planting out. They can also be established from tiller transplants.
New Zealand native grasses fall into two groups in respect to flowering. Grasses may flower every year (e.g. Fescue tussock; Lord 1998) or irregularly in mast-years (e.g. snow tussock spp; Connor 1996; Mark 1968). The mast flowering in snow tussock will be a problem with collecting large quantities of seed on a regular basis. An understanding of factors affecting the timing of mast years may allow the manipulation of seed production. Mark (1965a) suggested that the infrequent flowering years of snow tussock were dependent on abnormally high temperatures during the long-day period of summer in the season before flowering. However, more importantly flowering can be induced in snow tussock by burning (Mark 1965b; Payton & Mark 1979; Rowley 1970). Mark (1965b) reported that prolific flowering invariably occurred 15 months after a spring fire. Autumn burning also promoted flowering but the flowering response was delayed one complete season. The effect of burning appears to be due to the increased temperature in the blackened crowns during summer leading to prolific floral initiation in the autumn (Rowley 1970). In contrast to the first year after burning, the percentage of snow tussock flowering in the burnt plots was significantly less than in the unburnt plots in five of the six following seasons (Years 2-7).
Although burning is a possible management technique to ensure availability of seed, difficulty of obtaining resource consent and problems with limiting fire to the area required are potential limitations. Spring clipping of tussocks has also been shown to stimulate flowering of snow tussock, although the results are not as marked as burning (Mark 1965b; Rowley 1970). Mark (1965b) recorded 55-100% flowering after burning and 22-45% after clipping, while the values recorded by Rowley (1970) were 100% and 24%, respectively. Mark (1965b) also cautioned about the extrapolation of the clipping results as only one or two season’s results were available and there was variation between different sites.
In addition to increasing the number of snow tussock plants flowering, burning increased the number of florets per flowering tussock in the season after fire (Mark 1965b; Payton & Mark 1979). Thereafter, the mean number of inflorescences per flowering tussock declined, relative to numbers in unburnt plants, and was significantly lower by the 3rd year after burning. Seed set was higher in the burnt snow tussock but seed weights were significantly lighter than those produced by unburnt tussocks (Mark 1965b). This production of lighter seeds by burnt tussock may have implications for seedling establishment because of the demonstrated relationship between seed size and seedling vigour (Williams et al. 1968). Only limited information is available, but seed germination was lower in seed collected from recently burnt tussocks (Mark 1965b), in the one year studied.
Mark (1965a) found no consistent effect of altitude on the percentage of tussocks flowering but seed weight decreased with increasing altitude, and the small seeds produced above 1500m were almost all non-viable.
Mark (1965a) found that up to 7.5% of snow tussock florets contained unidentified gall midge larvae with eaten seed. In a five year study, White (1975) recorded that insect damage could cause 100% destruction of snow tussock seed in some areas. Three insects were recorded feeding in florets: Megacraspedus calamogonus, Dipltoxa neozandica and an unidentified gall midge. Insect damage to seed production in a range of snow tussock species varied from 0-100% in localised areas. He concluded that snow tussock species may not therefore have inherently poor seed set, as has sometimes been assumed, but despite their seasonal irregularity of flowering, seed production is always at risk of heavy attack by insects with very closely adapted host-specific biology. Sullivan & Kelly (2000) found that total seed predation was not significantly different in red tussock (C. rubra; 48%), midribbed snow tussock (C. pallens; 22%) and broad-leaved snow tussock (C. flavescens; 54%). In addition to predation caused by the three insects identified by White (1975), they also found ergot fungus at some sites, although only a few florets were infected. Seed predation significantly increased with decreasing altitude. For example, the six red tussock sites below 800 m elevation lost 90% of their florets to seed predation insects, while the six red tussock sites above 1000 m lost only 16% of florets. Insect predation has also been documented in fescue tussock (Lord & Kelly 1999). In contrast to the mast seeding snow tussock, healthy fescue tussock individuals flower and set seed every year. Total losses to pre-dispersal predation ranged from 46% to 96% of florets produced at lower altitudes, and from 9% to 44% at higher altitude sites. The seed predator responsible for most of the damage was not identified, but Cecidomyiid larvae and Diplotoxa were identified.
Espie & Lowther (1998) identified the potential problem of seed predation in restoration work. They reported low establishment when snow tussock seed was overdrilled or even sown into seed raising mix in the glasshouse. The seed, collected from a range of low altitude sites in Coastal Otago, had very low germination (1-3%) and high levels of insect damage.
Mark (1965) found that the germination of seed varied considerably between snow tussocks at a site, between sites and between years, ranging from 0 to 94%. There was no apparent relationship between seed size and viability, except that small seeds produced above 1500 m were almost all non-viable. The usual pattern of germination was an initial flush amounting to about 20% during the first 50 days, followed by intermittent germination over a period exceeding 4 years. Under laboratory condition, germination was slightly increased and accelerated by the removal of the glumes. Total germination was unaffected by light or by pre-treatment with continuous or intermittent low temperature. Pollock (1968) commented on the problems of poor seed set or filling and insect predation of snow tussock. He recommended stratifying seed for at least 3 weeks and up to 10 weeks for seed from the highest elevations. Molloy & Connor (1970) reported that germination of narrow-leaved snow tussock (C. rigida), red tussock and broad-leaved snow tussock was satisfactory when liberal quantities of seed were sown into soil in the glasshouse. Lowther (pers. comm.) found rapid germination of snow tussock seed sown in commercial seed raising mix in the glasshouse as long as seed with high laboratory viability was used.
Mark (1965) found snow tussock seedlings present on a range of tussock grassland sites, especially on a site at 1220 m on the Old Man Range that had been burnt 8 months previous. On this site, numbers reduced by about half after 3 years. Seedlings died principally during late summer and autumn, with those which had been lifted by frost in winter most vulnerable to surface desiccation. Grazing damage to seedlings was obvious, especially as they enlarged. However, only a few seedlings were killed by grazing in the first three years. On 10 montane-subalpine sites, Rose & Platt (1992) found that snow tussock seedlings were infrequent in areas subject to sheep grazing and they concluded that snow tussock was continuing to decline in abundance. There was a high abundance of seedlings and juveniles in areas retired from sheep grazing. Results from enclosures demonstrated that grazing by hares was capable of inhibiting snow tussock recovery. In areas retired form grazing, snow tussock seedlings were most frequent within 70 cm of mature tussocks and on microsites protected from frost-heave by short ground-tier vegetation
Direct drilling of native grasses has been investigated at the Tekapo Accommodation Reserve (Espie & Scott 1994). Laboratory germination of seed used varied: blue wheat grass (Elymus rectisetus) 74%; fescue tussock 32%; blue tussock 27%; silver tussock 14%; snow tussock. 0.5%; red tussock 0.3%. Early seedling establishment ranged from 11 blue tussock seedlings per m of row down to 3 red tussock seedlings per m. When expressed on the basis of viable seed sown the percentage establishment was 15% for blue wheat grass, 3% for silver and blue tussock, 2% for snow tussock, 1% for fescue tussock and 1% for red tussock. Longer term establishment was not assessed as severe grazing by rabbits resulted in a virtual establishment failure of all species (Espie pers. comm.). Al Shearer (pers. comm.) has had success in direct drilling silver and blue tussock, and to a lesser extent fescue tussock, but has obtained very low establishment from snow tussock. Espie & Lowther (1997) failed to find any seedlings after direct drilling snow tussock seed and attributed the failure to the use of snow tussock seed with very low germination.
Transplanting of nursery raised seedling has been widely used in revegetation of small areas of degraded or disturbed tussock grasslands (Nordmeyer pers. comm.; Pollock 1986). Normally these are well grown seedlings, often at least 2 years old, that are planted by hand in prepared areas.
The possibility of developing technologies, that are suitable for large scale mechanised planting of seedlings has been investigated by Espie and Lowther (1997; 1998; and unpub. results) for re-establishing snow tussock on land retired from farming. They planted 3 month old seedlings, raised in commercial potting mix, into a depleted browntop (Agrostis capillaries) sward. Seedlings were planted into the existing sward or into a furrow prepared by an experimental strip-seeder drill which removed a strip of vegetation 100 mm wide by 25 mm deep and cultivated the furrow to a depth of ca. 50 mm (Horrell et al 1991). Seedlings were sown with or without slow release fertiliser into areas that were unsprayed or sprayed with herbicide. Excellent establishment and survival of tussocks after 3 years (over 90%) was obtained where seedlings were planted in the drill furrow and herbicide applied. However tussock establishment was low (less than 35%) when seedlings were planted into the existing sward, even with herbicide. This was attributed to a rapid re-establishment of grasses, both from regrowth and from seed, in herbicided areas. The enhanced survival in the strip seeder furrow was attributed to improved control of competing vegetation. The application of slow release fertiliser to seedlings at planting had a large effect on growth (height and basal diameter) of snow tussock where competition from the existing vegetation was controlled. Caution must be used in extrapolating these findings as there was a virtual establishment failure when snow tussock seedlings were planted into ryegrass/white clover pasture that had been retired from grazing (Lowther unpub. results).
The response to slow release fertiliser agrees with conclusions of Molloy & Connor (1970) that snow tussock seedlings respond to fertiliser. They concluded that the major response was to phosphate but, this may be an artefact of their pot trial. In general, in undeveloped tussock grasslands the nutrient deficiency that most severely limits growth is nitrogen (Scott et al. 1995). Development of tussock grasslands is usually by introducing nitrogen-fixing legumes to build up soil nitrogen for the associated pasture grasses. Molloy & Connor (1970) found that foliage nitrogen levels were relatively high in snow tussock seedlings, even in the absence of fertiliser. They concluded that the nitrogen requirements of snow tussock may be quite small in comparison with fast-growing high-producing pasture grasses. Molloy & Connor considered, but dismissed, the possibility of mineralisation of organic matter releasing nitrogen for plant growth in the nil-fertiliser pots. However, enhanced mineralisation is known to occur in sieved soil in pots in the glasshouse and it is surprising that Molloy & Connor dismissed the possibility that mineralisation was releasing nitrogen for the tussock seedlings.
In view of the response of seedlings to fertiliser in both the field and laboratory, it is clear that fertiliser should be applied when seedlings are sown in low fertility soils. Although Molloy & Connor consider that the main response was to phosphorus and not to nitrogen, their conclusions do not appear to based on sound evidence. In view of nitrogen being the major deficiency in the majority of undeveloped tussock grassland soils, it is considered that nitrogen should be included in the nutrient mix applied at sowing. It would appear logical to continue to use a long term slow release fertiliser.
Direct planting of snow tussock fragments has been used by the Dunedin City Council to re-establish tussocks into an area that had been cultivated and sown but had reverted to browntop (N. Harwood pers. comm.). In spring, existing large snow tussock were dug up and split into medium sized fragments and planted the same day directly into the browntop. Survival was low, with less than 10% of plants surviving after 12 months. This result is not surprising as the recommendation is to plant the divided tillers into nursery soil for a year before planting out into the field (Pollock 1986). However, an alternative method is to soak the tussock fragments in water to initiate root development before transplanting (Espie & Lowther 1997). Over 70% of tussock fragments that had been soaked, to initiate roots, were alive after 4 years compared with a virtual failure where fragments were planted directly after splitting the original tussock. Cutting the fragments to different leaf lengths at planting had no significant effect on the percentage surviving.
Nutrient depletion, particularly nitrogen, through grazing and burning is recognised as a major problem in maintaining sustainable tussock grasslands (McIntosh 1997). Development of tussock grasslands for pastoral use depends on correction of nutrient deficiencies (phosphorus, sulphur and molybdenum) and the introduction of legumes to increase nitrogen through symbiotic nitrogen fixation (Scott et al. 1995). Restoration technology in degraded tussock catchments has involved correction of nutrient deficiencies and the sowing of leguminous plants (Nordmeyer pers. comm.).
On the basis of the classification that tussocks were species of slow growth habit (Moore 1956), O’Connor (1963) was not surprised by the lack of growth responses of snow tussock to fertiliser (nitrogen phosphorus, potassium and sulphur). However, these were short term tiller growth (1 month) measurements.
Fig 1. Glasshouse raised snow tussock seedlings reading for field planting
Fig 2. Snow tussock seedlings planted in drill furrow in herbicide treated area
Fig 3. Effect of competition on snow tussock. Top photo shows a snow tussock plant originally sown into drill furrow after herbicide treatment. Bottom photo shows a plant sown into existing sward without herbicide. Both plants sown with slow release fertiliser.
The effect on tussock growth from superphosphate application and sowing white clover and Maku lotus was demonstrated by Espie (1990) in the Waimakariri Basin. After 7 years, the average height of fescue tussock increased from 25 cm in the untreated area to 46 cm where fertiliser had been applied. Basal circumference had increased from 6 cm to 15 cm. In a glasshouse trial, nitrogen and phosphorus increased growth by nearly 4-times, with phosphorus alone growth was less than doubled, while nitrogen had little effect on its own. Espie concluded that on these high-country yellow brown earth soils, phosphorus and nitrogen are the most important soil nutrients influencing fescue tussock growth. On the basis of these results, Espie suggested that applying fertiliser and introducing legumes is a method of retaining or enhancing fescue tussock in degraded grasslands. However, the effect of grazing and exotic weed invasion and how to manage fescue tussock grasslands with the requirements of the associated plants was noted. O’Connor (1966) found a similar response to superphosphate and legumes with fescue tussock in a pastoral situation but only under a low grazing pressure and concluded that it is possible to retain/enhance fescue tussock in improved grasslands but only at the expense of reduced grazing pressure.
The 9-year trial of Scott (2001) in the Mackenzie Basin confirms the importance of fertiliser for tussock growth and the interaction with grazing, although he concluded that fertiliser effects on tussocks are slow. The trial was established in an area of Hieracium-infested fescue tussock grassland that had been ungrazed for 10 years. The application of nitrogen fertiliser significantly increased height of fescue tussock two years after initial application and the maximum response was reached 5 years after initial application. Height was higher after application of 23 kg nitrogen per ha than with 6 kg nitrogen. There was a significant effect of nitrogen on tussock flowering in years 5 and 6. Because of the ranking system used by Scott it is difficult to determine the effect of nitrogen on the abundance of tussock. The application of S-superphosphate also increased tussock height but the response was slow to develop. The slow response to S-superphosphate can be attributed to the time for adventitious legumes to establish and fix sufficient nitrogen to allow transfer to the tussocks. Available soil-nitrogen levels were low in the soil and the breakdown and transfer of nitrogen is known to be low in the absence of the grazing animal (O’Connor 1966). It is possible that the enhancement of tussock height by grazing (after an initial reduction) may have been due to increased cycling of nitrogen.
A key feature of the results for tussock restoration is the importance of nitrogen, either applied as fertiliser or built up in the soil by legumes, and the possible enhancement from some controlled grazing. Visual evidence of increased growth of snow tussock from legumes and fertiliser has been observed at Tara Hills (Boswell pers. comm.).
Applying fertiliser and introducing legumes would appear to potentially provide a cost effective means of improving tussock growth in depleted grasslands. However the effect of grazing and exotic weed invasion, and how to tussock management with the requirements of its associated plants need further study.
Fig 4.Experimentally enhanced growth of tussock with applied fertiliser, Mt John, Tekapo (David Scott pers comm.)
Duncan, R.P; Webster, R.J.; Jensen, C.A. 2000. Declining plant species richness in the tussock grasslands of Canterbury and Otago, South Island, New Zealand. New Zealand Journal of Ecology. 25: 35-47.
Espie, P.R. 1990. The impact of nutrients on fescue tussock grasslands. What’s new in Forest Research No. 189.
Espie, P.R.: Scott, D. 1994. Tekapo accommodation reserve ecological restoration. Interim Contract Report: March 1994. AgResearch, Invermay.
Espie, P.R.; Lowther, W.L. 1997: Deep Stream snow tussock restoration. DCC Contract Report CV/CON/12. AgResearch, Invermay.
Espie, P.R.; Lowther, W.L. 1998: Deep Stream snow tussock restoration. DCC Contract Report. AgResearch, Invermay
O’Conner, K.F. 1963. Studies on the management of snow-tussock grassland II. The effects of cutting and fertiliser on narrow leafed snow-tussock (Chinochloa rigida (Raoul) Zotov) at high-altitude sites in Canterbury, New Zealand. New Zealand Journal of Agricultural Research 6: 368-375.
O’Conner, K.F. 1966. The improvement and utilisation of tussock grasslands: a scientist’s viewpoint. Proceedings of the New Zealand Grassland Association 28: 59-78.
Lee, W.G.; Fenner, M.: Duncan, R.P. Pattern of natural regeneration of narrow-leaved snow tussock (Chinochloa rigida) ssp. rigida in Central Otago, New Zealand. New Zealand Journal of Botany 31: 117-125.
Lord, J.M.; Kelly, D. 1999. Seed production in Festuca novae-zelandiae: the effect of altitude and pre-dispersal predation. New Zealand Journal of Botany 37: 503-509.
Mark, A.F. 1965a. Flowering, seeding, and seedling establishment of narrow leaved snow tussock (Chinochloa rigida). New Zealand Journal of Botany 3: 180-93.
Mark, A.F. 1965b. Effects of management practices on narrow leaved snow tussock (Chinochloa rigida). New Zealand Journal of Botany 3: 300-319.
Molloy, D.P.B.; Connor, H.E. 1970. Seedling growth in four species of snow tussock (Graminaea). New Zealand Journal of Botany 8: 132-152.
McIntosh. P. 1997. Nutrient changes in tussock grasslands, South Island, New Zealand. Ambio 26:147-151.
Payton, I.J.; Mark, A.F. 1979. Long-term effects of burning on growth, flowering, and carbohydrate reserves in narrow-leaved snow tussock (Chinochloa rigida). New Zealand Journal of Botany 17:43-54.
Pollock, K.M. 1986. Plant materials handbook for soil conservation. Volume 3: Native plants. Water and Soil Miscellaneous Publications 95. Ministry of Works and Development, Wellington.
Rose, A.B.; Platt, K.H. 1992. Snow tussock (snow tussock) population responses to removal of sheep and European hares, Canterbury, New Zealand. New Zealand Journal of Botany 30: 373-382.
Rowley, J. 1970. Effects of burning and clipping on temperature, growth, and flowering of narrow-leaved snow tussock. New Zealand Journal of Botany 8: 264-282.
Scott, D. 2001. Fertiliser and grazing rejuvenation old fescue tussock grasslands. New Zealand Journal of Agricultural Research 44: 481-490.
Scott, D.; Maunsell, L.A.; Keoghan, J.M.; Allan, B.E.; Lowther, W.L.; Cossens, G.G. 1995. A guide to pastures and pasture species for the New Zealand high country. Grassland Research & Practice Series No. 4. New Zealand Grassland Association Incorporated, Palmerston North.
Sullivan, J.J.; Kelly, D. 2000. Why is mast seeding in Chinochloa rubra (Poaceae) most extreme where seed predation is lowest ? New Zealand Journal of Botany 38: 221-233.
White, E.G. 1975. An investigation and survey of insect damage affecting snow tussock seed production in some alpine tussock grasslands. New Zealand Journal of Agricultural Research 18:163-178.
Williams, W. A.; Black, J.N.; Donald, C.M. 1968. Effect of seed weight on the vegetative growth of competing annual Trifoliums. Crop Science 8: 660-663.
Wills, Barry 2002. Indigenous plants for pasture use. AgResearch, Alexandra.
This paper has been developed for this site and is reproduced with the permission of AgResearch. Your comments are welcomed and should be forwarded to Bill Lowther bill.lowther@agresearch.co.nz
| Page last updated on Wednesday, August 28, 2002 Site hosting by Landcare Research |
 |