Field experiments were carried out at seven sites in southern Australia from 2002 to 2006 to measure changes in plant frequency, root characteristics and summer activity for a range of grass and herb species or cultivars. Annual rainfall during the experimental period was on average 75 mm lower than the long-term average. Plant frequency differed significantly between species and between sites. Temperate grasses generally had higher frequencies than subtropical grasses, native grasses and herbs. Cocksfoot (Dactylis glomerata cvv. Currie, Porto), tall wheat grass (Thinopyrum ponticum cv. Dundas), winter-active tall fescue (Festuca arundinacea cvv. Fraydo, Resolute MaxP) and phalaris (Phalaris aquatica cvv. Atlas PG, Australian) were the most persistent of the temperate perennial species over the experimental period. The frequency of most cultivars declined from year 2 to year 4 after establishment, but the frequency of kikuyu (Pennisetum clandestinum cv.Whittet) and wallaby grass (Austrodanthonia richardsonii cv. Taranna) increased by over 5% from year 2 to year 3, and cocksfoot (cv. Currie) increased from year 3 to year 4. At two sites where measurements were made, there were significant differences in rooting depth between species. Whittet kikuyu was the deepest among all species with a rooting depth of up to 2 m, followed by phalaris, tall fescue, grazing brome (Bromus stamineus) and tall wheat grass. Root density was affected by plant genotype and soil structure. Root density of the species varied significantly in the subsoil (0.1–1.1 m) and deeper subsoil (1.1–2 m) but not in the topsoil (0–0.1 m). Green-leafiness over summer was generally higher for subtropical grasses, native grasses, herbs and some summer-active temperate grasses, than most temperate grasses with high summer dormancy.

Intercropping is considered by its advocates to be a sustainable, environmentally sound, and economically advantageous cropping system. Intercropping systems are complex, with non-uniform competition between the component species within the cropping cycle, typically leading to unequal relative yields making evaluation difficult. This paper is a review of the main existing metrics used in the scientific literature to assess intercropping systems. Their strengths and limitations are discussed. Robust metrics for characterising intercropping systems are proposed. A major limitation is that current metrics assume the same management level between intercropping and monocropping systems and do not consider differences in costs of production. Another drawback is that they assume the component crops in the mixture are of equal value. Moreover, in employing metrics, many studies have considered direct and private costs and benefits only, ignoring indirect and social costs and benefits of intercropping systems per se. Furthermore, production risk and growers’ risk preferences were often overlooked. In evaluating intercropping advantage using data from field trials, four metrics are recommended that collectively take into account all important differences in private costs and benefits between intercropping and monocropping systems, specifically the Land Equivalent Ratio, Yield Ratio, Value Ratio and Net Gross Margin.

To review pasture species for regions with 465–680 mm average annual rainfall, 22 perennial grasses and herbs were evaluated for pasture establishment and productivity in four states at seven locations where the arrest of groundwater recharge is considered necessary to ameliorate dryland salinity. Species represented introduced and native, temperate and subtropical grasses, chicory (Cichorium intybus L.) and plantain (Plantago lanceolata L.). This report describes establishment and yield; the following paper describes persistence and root characteristics. Yields were measured over 2–3 years except at one site, which suffered severe drought. Perennial ryegrass (Lolium perenne L., cv. Avalon) and tall fescue (Festuca arundinacea Schreb. = syn. Lolium arundinaceum. (Schreb.) Darbysh., cvv. AU Triumph and Resolute MaxP), cocksfoot (Dactylis glomerata L., cv. Porto) and phalaris (Phalaris aquatica L., cv. Holdfast and Australian) were the most productive species, with dry matter (DM) yields of 13.6–15.1 t/ha. For summer growth, Porto and Rhodes grass (Chloris gayana Kunth, cv. Katambora) were the most productive species; relative to Australian in summer, Porto and Katambora produced 41% and 26% more DM, respectively (95% confidence). Perennial ryegrass (cv. Avalon), tall fescue (cv. Resolute MaxP) and chicory (cv. Grouse) were particularly valuable for autumn growth; Avalon was 30% more productive than Australian. Tall fescue (cv. Resolute MaxP) was 32% more productive than Australian in winter. Avalon and AU Triumph were the most productive grasses and herbs in spring. Based on natural rainfall over the 2–3 years of measurement, the mean water use productivity, ignoring any runoff, was 10.5 kg DM/ha.mm for the three most productive species. Apart from kangaroo grass (Themeda triandra Forssk), native grasses gradually established, but over a prolonged period weeping grass (Microlaena stipoides (Labill.) R.Br., cv. Wakefield) was the most rapid. Perennial ryegrass, tall fescue, cocksfoot and phalaris maintained productive yields across a diverse range of soils and climates. Exploration of the diversity within these species in a nationally coordinated program of genetic improvement appears warranted for improving reliability and expanding the zone of adaptation.

This paper presents a case for the selection and development of a wider range of perennial grasses for pastoral use in the higher rainfall (annual rainfall >500 mm) zone of southern Australia, especially the southern sector of the Murray–Darling Basin. There is also a need to reconsider the use of ‘high-input’ pastures on hill lands by developing more appropriate recommendations for managing existing native grass pastures productively. Past experiments which compared native grass based pastures with sown pastures promoted the view that indigenous grasses were inferior in most respects to exotic improved species. Even though many of the findings were confounded with fertiliser, stocking rate, and other treatment effects, they reinforced the general direction of cultivar development programs which in the temperate zone have been based mainly on the 4 exotic C3 species Phalaris aquatica L., Dactylis glomerata L., Lolium L. spp., and Festuca elatior var. arundinacea (Schreb.) Hackel (syn. Festuca arundinacea Schreb). This has led to an imbalance in the adaptability and range of species available to be sown in pastures, particularly for sowing on less productive landscapes where stony, shallow, infertile, acid soils limit the persistence of current cultivars. The pre-European vegetation of temperate Australia comprised species with a capacity for active growth and transpiration during summer. The water use pattern resulted in soil moisture being near capacity in late winter and spring, and exhausted by summer’s end. Replacement of this vegetation with annual-growing and summer-dormant C3 species has changed the water use pattern so that soils are drier in spring and wetter in autumn. This has reduced the pre-winter soil moisture deficit, which in turn has increased rates of deep drainage in winter. Land degradation in southern Australia is a consequence of this changed water use pattern. Deep drainage of water beyond the reach of plant roots has mobilised salts stored in the landscape and caused watertables to rise, which has led to large areas becoming saline. Lack of growth in summer in pastures consisting of senescent annual-growing species and dormant C3 perennial grasses limits utilisation of the products of nitrogen mineralisation, which allows nitrate nitrogen to accumulate in summer and be readily leached by rainfall in autumn. This increases rates of soil acidification. Although there may be scope to reduce deep drainage by increasing pasture growth in spring in areas where there is little likelihood of summer rainfall, this is not the case in south-eastern Australia where significant falls of rain occur during summer and autumn.

This paper is the fifth in a series describing trials evaluating native and introduced grasses at eight locations across temperate Australia. In these trials, 62 perennial grass lines were assessed for herbage production, survival and recruitment under low fertiliser conditions using spaced plants produced in glass houses and transplanted into the field. Sites were grouped into three different climatic zones: Eastern Australian permanent pasture, Eastern Australian mixed farming and Mediterranean zone. For each of these zones, superior lines were identified and their potential use in permanent pastures or mixed farming discussed. Among the C3 grasses tested, several lines of Dactylis glomerata from France on the Mediterranean coast near the Spanish border and from north-west maritime France proved to be superior lines in all zones and were better than the standard comparator cv. Currie with regard to the attributes assessed. In general, the D. glomerata lines were superior to the C3 native species except with respect to survival and recruitment. The C4 introduced standard comparator Eragrostis curvula cv. Consol was outstanding with regard to its herbage production and survival in all climatic zones, although its recruitment was generally low. Native C4 lines of Themeda australis and Paspalidium jubiflorum yielded well in all climatic zones, and even produced more herbage than Consol in one zone. Survival rates of T. australis and P. jubiflorum were also very high but recruitment was low under the conditions of the trial. In mixed pastures C4 grasses may reduce fluctuations in feed supply as well as increase water use. As a consequence mixtures of superior C3 and C4 lines identified in this study are likely to be suitable for pastures on infertile soils and in zones subjected to continuous or periodic drought. Further work on the behaviour of these superior lines under sward conditions in association with forage legumes and their responses to grazing is needed; however, for the promising native lines, sward trials are not possible until suitable technology for commercial seed production and broad acre establishment of these lines has been developed.

The historical approach to pasture improvement in the high rainfall zone of temperate Australia has been to add introduced herbaceous legumes and to replace perennial native grasses with introduced species requiring high inputs of fertiliser for maintenance. The application of this high-input approach on land with low capability has lead to the loss of perennial grasses, erosion, soil acidification and increasing salinity on the lower slopes. This model of pasture improvement has not been successful on the margins of the wheat belt and in semi-arid regions. The Native and Low-input Grasses Network (NLIGN) was established in 1996 to coordinate research on grasses suitable for land with low capability and for semi-arid regions. The NLIGN multi-site evaluation project was initiated to test promising lines (accessions) of native and introduced grasses for low-input pastures at eight sites across southern Australia. The broad objective of the project was to identify native and/or introduced perennial grass lines that had possible commercial potential for low-input pastures. This objective was achieved in an initial 3-year evaluation phase, which began in 1998 and tested lines for persistence, production and palatability.