Pasture Management: Technology for the 21st Century

By CSIRO PUBLISHING

This book looks at current knowledge on management of pastures and rangelands for sheep production, of problems, of practical solutions where possible, and of priority areas for research. The areas considered extend from the high rainfall perennial pastures of south-east Australia and New Zealand, through the annual pasture, cropping zones to the semi-arid rangelands.

Pasture Management is the major reference on managing Australia's greatest natural resource: the resource which provides directly and indirectly a major part of Australia's export income.

• Language: English
• Publisher: CSIRO PUBLISHING
• Release date: Jan 1, 1994
• ISBN: 9780643106192

Book preview

Introduction

D.R. Kemp & D.L. Michalk

NSW Agriculture, Pasture Development Group.

Agricultural Research & Veterinary Centre

Forest Rd, Orange, New South Wales 2800

Grasslands are one of the world’s greatest resources, comprising nearly half of the earth’s land surface (Williams et al. 1968). Natural and sown grasslands provide most of the forage for the world’s ruminant and equine populations. Well managed pastures limit soil degradation and improve soil fertility. Over-population, though, is placing increasing pressure on grassland resources, with degradation and desertification becoming increasing problems in some parts of the world. It is evident that if the productive capacity of this global resource is to be sustained for future generations, management must be improved.

In Australia, pastures are one of our nation’s greatest natural assets and certainly Australian agriculture’s greatest resource. Native and improved pastures occupy some 60% of the land surface of Australia (approximately 460 m ha, excluding crops) and directly and indirectly generate income worth in excess of $12 billion per annum (Ockwell, 1990). Value adding provides substantial extra benefits to the national economy. Major industries such as wool ($3 billion direct to producers, 150 m sheep and 30% of the world’s wool production; Castles, 1989), cattle for meat ($3 billion and 19 m head) and dairy production ($1 billion and 3 m head) are based almost exclusively on pastures.

Pastures not only provide the basis for livestock production, but also improve soil fertility necessary to support the production of many crops and sustain the landscape. Grain production benefits substantially from a pasture phase in farming systems. In short, good pastures are the first line of defence against soil degradation in agricultural regions. Renewable pasture resources play a major part in sustaining Australia.

The history of pastures in Australia is a chequered one. In contrast to the American prairies and the African veldts, Australia’s native flora did not evolve under grazing by large groups of herbivores (Moore, 1970). Rather, native species were adapted to infrequent grazing by soft-footed macropods at low grazing pressures and, in the main, to low soil fertility and periodic burning. The introduction of cloven-hoofed animals in closely managed groups from the early 1800s, the failure of European settlers to appreciate the consequences of the regular pattern of droughts, and exploding rabbit populations, set in motion the process of deterioration of pastures and rangeland. Although changes in the pristine communities were recorded as early as 1853 (Bride, 1898), the full impact of degradation was not apparent until the 1890s when severe drought, livestock deaths and the demise of Australia’s pastoral resources, particularly in the rangelands area, attracted the attention of the press, governments, pastoralists and scientists. The pattern of changes documented in temperate woodlands of southeastern Australia (Figure 1) illustrates the impact of European settlement on Australian grasslands. Despite recognition of pasture degradation as a major national problem, it was not until the 1950s that widespread management changes were made which halted pasture decline in many parts of temperate Australia. The combined effect of the introduction of subterranean clover and companion grasses, the widespread use of phosphate fertilisers, the formulation of policies to restrict stock numbers in rangelands, the development of economic means of rabbit control, and the evolutionary responses within the native flora to livestock grazing, produced pastures that are now more productive and more sustainable than even a few decades ago, and certainly better than in 1900.

However, while many of the pastures sown with legume/perennial grass mixtures remain stable and productive over time, others have been invaded by a high proportion of less desirable species (Kemp and Dowling, 1991). Several factors are implicated in this more recent degradation of ‘improved’ pastures, including soil acidification, dryland salinity, decline in soil fertility, and poor grazing management practices. Reduced fertiliser usage and overgrazing are the consequences of declining profitability of livestock production as producers strive to survive the current cost-price squeeze.

Australian producers and scientists have a long history of utilising and adapting pasture resources and of developing technologies that have improved the productivity of that resource, often under difficult economic conditions. This process is a continuing one which demands periodic reviews of the state of knowledge, assessment of the direction of pasture research and identification of future avenues in pasture management that need to be supported. One of the most important obligations on Australians is the management of their pastoral resources for the health and welfare of the nation.

Figure 1: Progressive changes in the herbaceous communities of temperate woodlands in south-eastern Australia as a result of grazing by domestic herbivores.

Workshop and Review

The Australian Wool Corporation has been vitally concerned with management of the pastoral resource on which the wool industry is based. In 1990, the (now) Wool Research and Development Corporation (WRDC) allocated funds for a national review of pasture management research. The major part of that review was a workshop held in February 1992 at Armidale, New South Wales. Producers, scientists, farm advisers and agricultural administrators representing the major wool production zones of the southern half of Australia gathered at the workshop to review current knowledge and recommend future directions for research in pasture management. The invited papers, which reviewed important areas of pasture management, are published in this book, along with the outcomes of associated discussion sessions on research priorities. A major feature of this workshop was that one third of the papers were presented by producers. This maintained a realistic focus to the discussions and their outcomes.

Producers and scientists are now developing technologies for the 21st century that will ensure the viability of Australian pastoral industries. Producers are under considerable pressure to operate more efficiently within their existing resources by improving their management skills and refining their mechanisms for making decisions. No longer do they have the luxury to resow pastures with new species, apply unrestricted nutrients or control water supplies. Technology needs to be applicable in the short- to medium-term. Cost-effective technology must be based on a sound understanding of the scientific principles involved so that better and more reliable advice can be given. These general principles guided much of the discussion on research priorities. The priorities proposed apply to most of the grazing industries in temperate Australia.

Pasture Management and the Focus of this Book

What is pasture management? Agriculture by its nature is an interventionist activity. Pasture management is therefore the process of actively intervening in the production of plants and their utilisation by grazing animals to maintain or improve production while sustaining the resource. This definition includes plant as well as animal management and is somewhat different to that which would be applied to the management of non-agricultural ecosystems such as national parks, where the aim is to minimise intervention and to limit the influence of external factors. From an ecosystem perspective, pastures are not considered to be ‘climax’ communities which will return to their current state when grazing pressure is removed, but rather part of a ‘transition’ ecosystem where they may exist in one of many ‘states’ (Westoby et al. 1989) with a trend that is often towards less desirable agricultural ecosystems.

This retrogressive trend is a consequence of a lack of species diversity. Pastures often comprise a limited number of herbaceous annual and perennial species; this tends to increase the instability of such ecosystems. The high variability of the Australian climate means that most pastures are compromises which aim to survive the ‘average’ season. In contrast, trees and shrubs typically have to survive the worst seasons. The introduction of livestock on which the producer depends for his livelihood, adds an additional complicating dimension. These factors mean that pastures need active management to remain productive and sustainable within the constraints set by climate and society.

Pasture management can therefore comprise many activities. The challenge in reviewing pasture management was to identify those activities which can have a significant impact but where knowledge may be limited, so that future research can be directed into priority areas. Considerable previous work and several earlier reviews (e.g. Leigh and Noble, 1972; Wheeler, Pearson and Robards, 1987) had emphasised animal management, often within grazing systems that did not allow much flexibility for pasture management in the ways practised by many producers. Management of plants and of the grazing process had received limited attention, apart from a scientific review of pasture ecosystems, some 15 years ago, which covered the principles of plant interactions in pastures (Wilson, 1978). Previous reviews and grazing management research had often assumed that plants would always be there, or could easily be replaced by resowing. However, there is now a clear appreciation that this is not always the case, and that many of the principles of ecosystem management which were dismissed as applicable only to rangelands are seen as having relevance for higher rainfall pastures as well. Plants do need to be properly managed to remain productive within pasture ecosystems. Management of plants and the grazing process then became an important focus of this review.

Once this focus was established, important scientists in the relevant fields were invited to prepare a review for presentation and discussion at the workshop and publication in this book. It was also known that exciting new practices that could lead to more efficient production systems were being developed on farms. Innovative producers were asked to prepare papers on their practices. The last paper was also given by a producer to bring the scientific discussions into the perspective of a farm. The practices used by these producers and the research being undertaken in grazing management was of such interest that a tour of those speakers was arranged through six centres in NSW and Victoria in conjunction with the respective Grassland Societies. Some 600 producers attended those meetings, indicating the keen interest within the grazing communities for new information and better technologies. The Grassland Societies have made pasture management a major theme of their current programs.

The present review is timely in that pasture research within Australia has suffered a decline in resources and personnel since the late 1960s. This decline is in common with a general lack of support for science within many western countries. Concern at this decline by institutions such as the WRDC, other rural industry research and development organisations, state departments of agriculture and universities was an additional reason for this review. It was timely to review and regroup resources to maximise the efficiency and impact of pasture research and development. Pastoral industries supported this process and have helped considerably in focusing research at an appropriate level, as evidenced by the outcome of research discussions presented in this book. Research programs are now much more driven by industry than in the past and this change is generally supported.

In this book the state of knowledge about pasture management is first considered by practicing scientists. The resource base, management of pasture establishment and of nutrients, grazing systems for the perennial pasture, annual pasture and rangeland zones (supported by a review of the innovative practices being followed in New Zealand), and ways of integrating animal and plant management on the farm are each reviewed. The production systems used by innovative producers and the problems they encounter in each of the major zones are then considered, along with a producer’s view of the over-riding problems in integrating pasture management on the farm. These chapters are followed by a summary of the outcomes from workshop discussions on the priorities for pasture management research and a discussion on where we should go from here.

The papers presented in this book show that pasture research is alive and well. Past work has been considered and digested. The outcomes are a refocusing on those aspects of pasture management considered to offer significant gains for pastoral agriculture into the early 21st century. When pastoral practices are reviewed in the next century we would confidently expect that they will be different in important ways from those commonly used at present. The work considered in this book will have had a valuable impact on those changes

References

Bride, T.F. (1898). ‘Letters from Victorian Pioneers’. (Melbourne.)

Castles, I. (1989). ‘Year Book Australia 1988’, p. 72. (Australian Bureau of Statistics: Canberra)

Kemp, D.R, and Dowling, P.M. (1991). Species distribution within improved pastures over central NSW in relation to rainfall and altitude. Australian Journal of Agricultural Research, 42, 647-59.

Leigh, J.H., and Noble, J.C. (Eds) (1972). ‘Plants for Sheep in Australia’. (Angus & Robertson: Sydney).

Moore, R.M. (1970). Australian Grasslands. In ‘Australian Grasslands’ (Ed. R. Milton Moore) pp. 87-100 (Australian National University Press: Canberra).

Ockwell, A. (1990). In ‘Australia and the Australian Economy’, (Ed. D.B. Williams), pp 27-49, (Sydeney University Press; Sydney).

Westoby, M., Walker, B.H., and Noy-Meir, I. (1989). Opportunistic management for rangelands not at equilibrium. Journal of Range Management, 42, 266-73.

Wheeler, J.L., Pearson C.J. and Robards, G.E. (Eds) (1987). Temperate Pastures. Their production, use and management’, (Australian Wool Corporation / CSIRO: Melbourne).

Williams, RE., Allred, B.W., DeNio, RM. and Paulsen H.E.(1968) Conservations development and use of the world’s rangelands. Journal of Range Management, 21, 355-60.

Wilson, J.R. (Ed.) (1978) ‘Plant Relations in Pastures’ (CSIRO: Melbourne).

PASTURE MANAGEMENT: STATUS, ESTABLISHMENT AND NUTRIENTS

The Pasture Resource Base: Status and Issues

A.D. Wilson¹ and R.J. Simpson

Faculty of Agriculture and Forestry,

The University of Melbourne, Parkville, Victoria 3052

¹ Present address: ‘Cal Col,’ Deniliquin, New South Wales 2710

Summary

This paper presents an overview of the pastures present on high-rainfall perennial, annual crop zone and rangeland farms in the sheep grazing areas of Australia, as a background to the associated papers on their management. Information on the pasture compositions found on farms is summarised and the targets for management are discussed. A number of the more prominent pasture management issues are analysed.

In high-rainfall areas, annual volunteer grasses and broad-leaved weeds make up higher proportions of pastures than sown clovers and perennial grasses and there are significant invasions of weedy perennials of low productivity. Clover content is generally low and a significant proportion of pastures contain no clover, despite its known contribution to soil fertility and animal production. Also, the older varieties of subterranean clover and perennial ryegrass, which have poorer winter growth, remain dominant.

In the crop zone, pastures are almost wholly annuals, with a dominance of species of short growing season and low dry-season quality, such as Vulpia spp. The desired composition is driven more by the demands of the crop industry than the livestock industries, but both could benefit from the greater use of perennial species, such as lucerne, which provides a longer growing season and contributes nitrogen to the soil.

In the rangelands, native species remain dominant, but levels of more palatable perennials have been depleted and need restoration. Other needs are to reduce the levels of unpalatable shrubs, to determine optimum rates of pasture utilisation and to develop improved grazing strategies.

There is considerable scope for improvement in our understanding of the ecological management of mature pasture systems, and in the application of new technologies in pasture renovation.

Introduction

The research needs for pastures in the sheep grazing areas of southern Australia have recently been reviewed by Wheeler et al. (1987). They noted the numerous problems of pasture pests, poor seed supply, establishment difficulties, plant nutrition, rhizobia, weeds, acid soils, short crop rotations and poor grazing management in improved pastures, with a general conclusion that pastures were far from ideal. A similar conclusion has been reached for native pastures of the semi-arid rangelands, where composition has been changed dramatically by grazing and there is now a dominance of more ephemeral and wiry grasses, to the detriment of land stability and sheep production (Dowling and Garden 1991).

Most of the literature on pastures concerns the ‘ideal’ pastures that should be sown, or the pastures which are present on research stations soon after those mixtures have been sown. However, pastures on farms rarely conform to these ideals. This should not be surprising since statistical information suggests that, in Victoria for instance, only about 4% of pastures are resown in any one year (K. Bishop, pers. comm.). There are thus large areas of pastures that are at least 25 years old, which might now be termed naturalised pastures, rather than sown pastures. In these pastures natural selection will most likely place an emphasis on species survival, rather than sward production or high quality. If compositions other than this ‘survival mix’ are desired, pasture management will be a continual requirement. The target should be to manage pasture composition towards the ideal, even though that ideal may not be considered stable and is perhaps rarely reached.

It is valuable to examine the sparse information that we do have about pasture composition and production on farms and to see what it tells about research and management requirements. Coupled with this is the need to look closely at the ideal pasture that we want as a target of our management. Production and quality objectives are understood, but it may be difficult to optimise for both production and quality at the same time. Furthermore, a farmer may also be more interested in other objectives than production, such as easy care pastures, the survival of favoured species, or freedom from vegetable fault.

As a background to the following papers on management, this paper examines the current composition of pastures on farms, and the causes of the less than ideal mixtures. It also examines what types of targets we may have for those pastures, given the needs of livestock and maintenance of the land resource. Pastures are considered within the three main categories, namely high-rainfall perennial pastures, annual crop zone pastures and rangeland pastures. These zones are not always distinct, but since these divisions have a profound influence on composition and management issues, they form a suitable classification for this discussion.

High-Rainfall Zone Pastures

Composition

Pastures in the high-rainfall areas have changed dramatically from the original native pastures and some change is still occurring, although at a slower pace. The original native pastures are thought to have been dominated by tall, warm-season perennials, such as Themeda australis, and to have contained a large number of species (Whalley et al. 1978). With the introduction of grazing they changed to pastures of short perennials, such as Danthonia spp. and Bothriochloa, but they still contained a large number of grass and forb species, although without significant legumes. They presented a sparse cover with poor production in winter and low soil protection (Lang et al. 1952). There was some initial research aimed at improving these by grazing rotations, but this was largely overtaken by the ‘sub and super’ revolution.

Introduced perennial grasses were components of these pastures, but not essential ones. Lang et al. (1952), in their survey of pastures in the Western District of Victoria, recorded about 40% of pastures as improved at that time, and of these 7% had a ‘problem’ of clover dominance arising from the sowing of subterranean clover (Trifolium subterraneum) without grasses. The key change, therefore, was the introduction of legumes and the fertilisers needed to support those legumes. There are now numerous reports of clover decline, but it is a long time since anyone has reported a problem of clover dominance.

The more recent developments have been towards the development of species and varieties that fit into niche environments, such as sainfoin (Onobrychis viciifolia) for deep sands, or that meet more specific needs, such as Persian clover (Trifolium resupinatum) for hay crops and other species for winter growth or improved persistence (Reed 1987).

A summary of surveys of high-rainfall pastures is presented in Table 1. In one of the more comprehensive recent surveys, pastures in the Western District of Victoria (500-800 mm rainfall) were found to contain an average of 12% subterranean clover, 15% improved grasses, 58% volunteer grasses and 15% broad-leaved weeds (Quigley et al. 1993). A wider survey of 300 sites recorded a median of about 20% subterranean clover and about 15% improved perennial grasses, with improved perennial grasses absent from about 25% of sites. Only 25% of pastures contained 30% or more subterranean clover and only 15% contained 30% or more improved grasses (Quigley 1990) (see Figure 1). In a higher rainfall zone (750-950 mm), the median legume and exotic grass contents were again about 20% each, with the most prevalent species being volunteer perennials such as Yorkshire fog (Holcus lanatus), sweet vernal grass (Anthoxanthum odoratum), crested dogstail (Cynosurus cristatus) and browntop bent grass (Agrostis capillaris) (Ward and Quigley 1992).

A similar position has been recorded in New South Wales. According to a survey by Kemp and Dowling (1991) (see Table 2), the broad composition of pastures on the central Slopes and Tablelands of NSW was 42% legumes, such as subterranean clover (which may have been overestimated by the point quadrat method used), 36% annual grasses such as soft brome (Bromus mollis), vulpia (Vulpia spp.), barley grass (Hordeum leporinum) and annual ryegrass (Lolium rigidurri), and 16% perennial grasses (mainly introduced). In all, more than 65 species were recorded including many weeds and 10 native grasses. Perennial species were found down to 600 mm rainfall, but only exceeded the proportion of annuals when rainfall exceeded 800 mm for legumes and 900 mm for grasses. It was concluded that each group, perennials and annuals, grasses and legumes, occupies a different niche in the pasture and the groups do not replace one another exclusively.

Both Quigley et al. (1990) and Kemp and Dowling (1991) concluded that there is potential to improve these pastures by sowing better varieties of grasses and by increasing the proportion of clover. The proportion of clover is regarded as being below that required for optimum animal production. Problems of persistence of legumes in pastures have been perceived for some time (Carter et al. 1982; Gramshaw et al. 1989; Reed et al. 1989; Anon n.d.). Unfortunately, there is a general lack of quantitative evidence of the extent and nature of decline or of the trends in pasture productivity. Perception of declining legume composition in pastures is more or less based on memories of productive clover-based pastures during the ‘sub and super’ boom of the 1940s and 1950s which contrast with the low proportions of legume recorded in recent pasture surveys. Nevertheless, even in the face of mounting evidence that pastures no longer conform to our ideal of high legume content and are often not dominated by ‘desirable’ legume and grass species, there is still scepticism regarding the extent or seriousness of the problem. This is undoubtedly because quantitative evidence is lacking and year-to-year variation in climate regularly tips the ecological balance in favour of, or against, high legume content in pastures (Rossiter 1966).

Table 2. Mean proportion of species in sites where they were present in central New South Wales pastures (from Kemp and Dowling, 1991).

Figure l. The proportion of subterranean clover and sown perennial grasses (Lolium perenne, Phalaris aquatica, Festuca arundinacea and Dactylis glomerata) in pasture swards of 300 paddocks examined in spring 1989 by Quigley et al. (1990). There was less than 20% subterranean clover and less than 15% sown perennial grasses on about half the paddocks.

This scenario reinforces our concern that there are no regular and standardised surveys of the state of pastures. The only consistent long-term data base is the Australian Bureau of Statistics. This data base is used to show long-term trends in production, management practice and stocking rates. However, many critical details (e.g. pasture composition, soil fertility, etc.) desirable for analysis of emerging pastoral problems are not able to be collected by the Bureau. Data concerning pasture condition that have been collected must be reported in a manner that will ensure they can be accessed in future years. Despite the existence of excellent and suitable scientific journals in Australia, it is still common for survey information to be reported only in annual reports or conference proceedings, without complete accounts of materials, methods or locations.

Despite the paucity of information, it is clear that annual grasses are a prominent and permanent part of pastures in the high-rainfall zone and that it is difficult to achieve the higher proportion of clover thought to be desirable.

Figure 2. Fluctuations in the percentage dry weight of subterranean clover from year to year at three sites near Kojonup, Western Australia. A major part of the variation was attributed to the distribution of rainfall in autumn, together with the observation that increased grazing pressure reduced fluctuations (from Rossiter 1966).

Given the widespread nature of these findings, it is difficult to avoid the conclusion that the so-called weedy annual and perennial grasses are a natural feature of ‘improved’ pastures. In many places this has been attributed to ‘clover decline,’ which may well be a component, but it could equally be attributed to a natural divergence of species composition in long-established pastures. Whilst only a few species are sown, pastures now contain a larger number of species, as found in the original native pastures they replaced.

Management objectives

Composition has an important influence on the feeding value of pasture and affects yield, quality and the length of the growing period. There is a common assumption that the desirable pasture will contain a high proportion of sown perennial and legume species, yet the data to support this assumption are sparse and contradictory. What is the management target?

The main limitations to the value of medium- to high-rainfall pastures in southern Australia are seen to be the low availability of forage early in the growing season (late autumn and early winter) and the low nutritive value in the dry season (Doyle et al. 1989). Winter growth is often seen as a problem. For early-season growth, it is generally recognised that grasses are superior to clovers. There is also evidence that perennial grasses do not produce as well as annual grasses early in the growing season (Doyle et al. 1989). On the other hand, perennials have a longer growing season and hence a longer period of the year in which green feed is present.

In terms of quality it is well known that the weight gain of stock grazing on legumes, such as lucerne, perennial clovers and annual clovers, is greater than on grass-based pastures (Reed et al. 1972; Freer and Jones 1984), because of higher forage intake and more efficient metabolisation of energy and protein. Smaller differences also occur in quality between grasses, but these are probably of less consequence than differences in persistence. For instance, phalaris (Phalaris aquatica) has been found to be superior to perennial ryegrass (Lolium perenne) and cocksfoot (Dactylis glomerata) for lamb weight gain and weight gain by ewes in winter on the Southern Tablelands (Axelsen and Morley 1968).

Out-of-season growth is also important for quality and can be achieved by including summer-growing perennials, such as lucerne (Medicago sativa), in winter pastures, or year-long green perennials in summer pastures (Lodge and Whalley 1983). In many of our grazing lands, the dry season is dry in a relative sense, but is not rainless. This is particularly so in New South Wales. Hence significant growth can be obtained from perennials such as lucerne, which extend the growing season into summer (Kenny and Reed 1984; Donnelly et al. 1985). However, lucerne is not often a component of mixed pastures.

Progress is slowly being made in addressing the limitation imposed on animal production by poor pasture growth in winter. Sirolan and Sirosa phalaris, Haifa white clover (Trifolium repens) and Karridale subterranean clover are all examples of cultivars that have outclassed other cultivars in this respect (Reed 1987; Clark and Hirth 1987; Clark et al. 1991). It must be admitted that change has been slow. In Victoria, for instance, Mt. Barker subterranean clover, which is a notoriously poor producer in autumn and winter, was the main mid-season cultivar from 1906 until 1988, when Karridale became available through the National Subterranean Clover Improvement Program. Moreover, Reed (1987) has argued that much more improvement is yet to be made in winter growth of clover and many alternative grass and legume species, by exploiting the genetic variation of Mediterranean ecotypes.

In addition to these individual objectives, we must also consider how they are to be combined. This may not be simple because high quality is to some extent associated with lower yield. For instance, whilst clovers support superior growth by sheep late in the season, annual grasses such as annual ryegrass (Lolium rigidum) and large seeded species such as oats (Avena sativa) and barley (Hordeum vulgare), exhibit higher growth rates than clovers and perennial grasses early in the growing season (Doyle et al. 1989). Furthermore, the response obtained for wool production may be different from that for weight gain. Hence, it may not always be clear which pasture composition is the most desirable.

The problem is illustrated by reference to alternative phalaris- and lucerne-based pastures on the Southern Tablelands of New South Wales, which were investigated in a comprehensive grazing trial by Donnelly et al. (1983, 1985). They concluded that lucerne-based pastures were superior to phalaris for lamb and weaner growth, because lamb growth rate was higher on lucerne and the pasture had a longer growing season. On the other hand, phalaris-based pastures were superior for wool-growing sheep, because they were resistant to higher stocking rates, tolerant of water-logging and drought, and required less management. Different outcomes are likely in other environments and with other species, but it is apparent that the best pasture for a given site is partly dependent on the purpose for which it is to be used, with an emphasis on quality where the product is weight gain and an emphasis on carrying capacity where the product is wool.

From the foregoing discussion, it is apparent that no single species possesses all the desirable attributes for a pasture. A mixture of species— legumes or forbs for quality and nitrogen fixation, annual species for early growth, perennial species for extending the growing season and grasses for forage persistence over the dry season— is still the target. The emphasis given to any one of these attributes, and hence the desired composition, will vary from region to region.

In some regions high quality and carrying capacity will be met by the one mix of pasture species. However, in others the best objective for a sheep farm as a whole could be different species in different paddocks, particularly where there is a variety of land types and species that favour those types, with an emphasis on quality in some and quantity in others. This would allow joint objectives, such as high quality for weaner growth and high carrying capacity for wool growth by dry sheep to be both met on the one farm. This is not a common practice, except in the case of lucerne paddocks.

Annual Crop Zone Pastures

Composition

In the cropping zone, pastures become a subsidiary to crop production, despite occupying up to two thirds of the land at any one time. Management may thus not be directed at improving animal production, although there is often a common goal in higher legume content. Pastures are only short-term and the values of these pastures are mostly seen in terms of their roles in increasing soil fertility and as disease breaks for the ensuing cereal crop. Hence, little attention is given to their value for sheep. When viewed in these terms, a high proportion of annual legume is desired, despite feed shortages in autumn. However, pastures with a very high proportion of legumes are not stable and can only be maintained by chemical application. Pastures tend to follow a pattern of high legume content initially with an increase in broad-leaved weeds and grass with time.

There is little survey information on current composition. Information from two limited surveys (see Table 3) shows that clover content is often less than desirable and that weedy annual species are often dominant. In the northwest of Victoria, R. Latta (pers. comm.) estimates that whilst some pastures have a high annual medic content, 50% of the area could be classified as ‘unimproved’, with an estimated composition of 60% annual grasses [barley grass, brome grass (Bromus spp.), vulpia and annual ryegrass], 10% annual legume and 30% broad-leaved weeds [capeweed (Arctotheca calendula), onion weed (Asphodelus fistulosis) and brassicas (Brassica spp.)].

The high incidence of vulpia is regarded as a major problem in New South Wales pastures because of its low productivity and low feed value (Leys et al. 1991). This species has probably increased in recent years because of the increase in grazing intensities and the application of herbicides in the crop phase that eliminate other annual grasses, but not vulpia. A common method of managing pasture composition is to apply non-selective herbicides, such as glyphosate or paraquat. However, much improvement in pasture composition can be achieved by grazing management alone (Doyle et al. 1992). Pastures in the annual crop zone are often not fertilised directly and little, if any, grazing management is practised.

In central and northern New South Wales, where rainfall is non-seasonal, the original native pastures contained summer perennial grasses, with the addition of naturalised annual grasses and legumes. With cultivation, the native perennials have largely been eliminated and replaced with volunteer and sown annuals. Lucerne may be included and is important in lifting stocking rates (C.L. Mullen, pers. comm.). It is also important for raising the yield of subsequent cereal crops (Holford 1980). In the marginal cropping zone of northern New South Wales, there is often no replanting of pastures, and pastures consist of barley grass, brassicas, vines [such as tar vine (Boerhavia diffusa) and bindweed (Convolvulus arvensis)], rolypoly (Salsola kali) and naturalised medics, which have poor carrying capacity over the summer in comparison with the original perennial grass/annual legume pastures (A. Bowman, pers. comm.). In these areas, native perennials may be recommended in preference to introduced perennials, not necessarily because of their yield, but because of their drought tolerance and persistence. Persistence is a characteristic that is often ignored in agronomic trials.

Composition is difficult to specify in annual pastures, with large differences between years according to rainfall. Instability in botanical composition is a natural feature of annual pastures, with wide swings in composition in relation to climate, fertiliser history, grazing management and fodder conservation (Rossiter 1966). The extremes are illustrated by the observations of White et al. (1980) who recorded 96% annual grasses during a drought year and 81% subterranean clover in the following year. Years of early opening rainfall are characterised by a high proportion of legume, years of late rainfall by a high proportion of grasses and years with early rainfall followed by dry period before winter, by a high proportion of capeweed and Erodium spp. (Rossiter 1966).

Stocking rate also has an influence with broad-leaved weeds, such as capeweed and Erodium, tending to increase at higher stocking rates (White et al. 1980). Increasing the stocking rate leads to an increase in vulpia and a decline in annual ryegrass (Rossiter 1966, Sharkey et al. 1964). The evidence for effects of higher stocking rates on subterranean clover content are equivocal, with some reports of a decline at higher stocking rates (Sharkey et al. 1964) and others of no effect (Rossiter 1966) or even an increase (Curtis et al. 1989). Curtis et al. (1989) state that sheep appear to graze grasses in preference to subterranean clover, whereas Curll (1982) says sheep select clovers. Clearly, there is no general rule, with selection of clover varying with the relative palatability of the companion grass.

Composition also varies with pasture age. There is a well established succession from annual legumes to grasses and subsequently to weeds as a pasture ages (Rossiter 1966). The sequence given by Rossiter for Western Australian annual pastures over 7 years was Trifolium spp. 70%, Trifolium + Vulpia, Erodium, Erodium + Trifolium, Vulpia + Erodium, Echium 50% and Echium 75%.

The most common feature of annual crop zone pastures is the almost complete lack of management. Composition varies widely with the age of the pasture and the characteristics of autumn rainfall and, even where summer rainfall is common, little attention is given to including perennials in pastures.

Management Objectives

Many of the conclusions reached for high-rainfall pastures also apply to annual and crop zone pastures. The main deficiencies of pastures in the annual crop zone are pasture quality during the summer and pasture quantity during the autumn (Doyle et al. 1989). Carrying capacity and pasture quality are both high during the growing season, but poor feeding value leading to liveweight loss and under-nutrition becomes a major problem during summer. Improvement can be obtained by paying attention to either dry season quality or to changing species to obtain some green feed during this period. Legume content is again important for pasture quality, but the value of quality to the farmer will depend on the type of sheep grazed, as it does in the high-rainfall zone. Quality is important for weight gain, but much less relevant for wool-growing sheep.

The value of dry forage is sharply influenced by species composition, with relatively poor value for many of the volunteer grasses. Clovers and medics are important components of quality during the dry season (Thorn and Perry 1987). The data suggest that dry capeweed, annual ryegrass and soft brome are poorer than subterranean clover residues, whilst ripgut brome (Bromus diandrus), vulpia, barley grass and Erodium residues are poorer still (Purser 1981). On the other hand, the dry residues of grasses are more resistant to weathering and trampling loss than those of clovers and forbs. The objective therefore is to maintain a high legume content, together with grasses that have higher quality as dry forage. Vulpia and Erodium are undesirable.

An important alternative to improving the composition of annuals is to increase the length of the growing season by including perennials. Particular evidence for this is the response that can be obtained by adding lucerne to the pasture mixture. At Horsham, T. Young (pers. comm.) estimates that farms with lucerne in their pastures carry 60% more sheep than farms without lucerne. At Charlton (430 mm rainfall,