Experiences with Farm Pastures at the Former CSIRO Samford Research Station, South-East Queensland, and How These Relate to Results from 40 Years of Research
Total Page：16
File Type：pdf, Size：1020Kb
Tropical Grasslands (2003) Volume 37, 151–164 151 Experiences with farm pastures at the former CSIRO Samford Research Station, south-east Queensland, and how these relate to results from 40 years of research R.M. JONES AND G.A. BUNCH Introduction CSIRO Livestock Industries, Brisbane, Australia Samford Research Station in coastal south-east Queensland (27°22′S, 152°53′E) was established Abstract as a field station for the former CSIRO Division of Tropical Pastures. Essentially it comprised 2 dairy farms, purchased in the late 1950s. The Samford Research Station in coastal south-east station was extensively used for research during Queensland was operated by CSIRO from the the 1960s and 1970s, but the requirement for late 1950s–2002. Many experiments on tropical research then gradually declined and it was legumes and grasses were carried out on the closed in 2002. The station always supported station and many farm pastures were sown. several hundred cattle, primarily run on farm Furthermore, once most experiments had ended, pastures. they were grazed as farm pastures. Hence, con- Consequently, considerable experience was siderable experience was gained about how dif- gained with pasture species under farm grazing ferent species did or did not persist under farm over the past 40 years. In some cases, these spe- grazing. This paper outlines these experiences cies had been sown as farm pastures and in other with individual species of legumes and grasses instances they were “left-over” after experiments and discusses how they relate to experimental had ceased. In this paper, we relate our experi- results. In general, performance of species in ences with pasture species under farm grazing farm pastures was similar to the results from and consider how this relates to research results, experiments, provided the experiments were run with particular emphasis on results from for long enough and with appropriate grazing Samford. We then discuss some possible reasons pressures. The main reason for the failure of for any differences between “farm experience” short-term experiments to predict long-term and “research results”. One of us (R.M. Jones) results was that they did not allow sufficient time was involved in experiments at Samford for many to establish whether recruitment could compen- years between 1969–2000 while G.A. Bunch sate for death of the original plants. worked as a field technician at Samford from The most persistent legumes were perennial 1969–1993 and was Samford farm manager from peanut (Arachis glabrata), Pinto peanut (Arachis 1993–2002. pinto) cv. Amarillo and leucaena (Leucaena leucocephala), followed by siratro (Macroptilium Samford Research Station atropurpureum) cv. Siratro, white clover (Trifo- lium repens) and Shaw creeping vigna (Vigna Soils and vegetation parkeri), all of which persisted in some situa- The farm comprised 290 ha, with approximately tions. The most persistent and aggressive grass 10% being well drained alluvial flats and 30% was Bahia grass (Paspalum notatum), with gleyed soils bordering the well drained soils or reasonable persistence from Queensland blue along small drainage lines. All these areas had couch (Digitaria didactyla) and setaria (Setaria been cleared prior to development of the station sphacelata). and the dominant species would have included carpet grass (Axonopus fissifolius syn. A. affinis), green couch (Cynodon dactylon), paspalum Correspondence: Mr R.M. Jones, 20 Fairholme Street, Kenmore, Qld 4069, Australia. (Paspalum dilatatum) and blue couch (Digitaria e-mail: [email protected] didactyla). 152 R.M. Jones and G.A. Bunch A further 40% of the farm was comprised of although several areas were re-sown, mainly in red-yellow podsolic soils with small areas of the early 1990s. The early sowings were usually prairie soils on gently sloping hillsides. Most of heavily fertilised with superphosphate plus trace these areas had been previously cleared and sup- elements at 600 kg/ha and potassium chloride at ported similar grasses to the alluvial flats, along 250 kg/ha as a basal fertiliser. During most of the with blady grass (Imperata cylindrica). Other 1960s, 250 kg/ha of superphosphate and areas were cleared and sown after 1960. The 125 kg/ha of potassium chloride were usually dominant tree genera prior to clearing were applied annually as maintenance fertiliser. After Eucalpytus and Tristania spp. The remaining that, levels of fertiliser were reduced and typi- 20% of the farm was comprised of lithosols on cally 250 kg/ha of superphosphate, sometimes hilltops. The small areas of lithosols that had including Mo, was applied every 2–3 years. been cleared prior to 1960 supported kangaroo Many species were included in farm sowings, grass (Themeda triandra) along with some of the but the main legumes were siratro (Macroptilium grasses previously listed. Further clearing of atropurpureum) cv. Siratro, greenleaf desmodium these lithosols was carried out in the 1960s while (Desmodium intortum) cv. Greenleaf, silverleaf some areas were left uncleared. desmodium (D. uncinatum) cv. Silverleaf, Average annual rainfall was 1100 mm, two- leucaena (Leucaena leucocephala) and white thirds of which fell in the 6 warmer months from clover (Trifolium repens). Smaller areas were October–March. Frosting of varying intensity sown to Kenya clover (Trifolium semipilosum) was recorded on the alluvial flats, at about 40 m cv. Safari, Pinto peanut (Arachis pintoi) cv. above sea level, in every year and there was usu- Amarillo, lotononis (Lotononis bainesii) cv. ally lighter frosting on the low–intermediate Miles, greater lotus (Lotus uliginosus) cv. Maku, hillslopes. The frosting records given by Cook round-leafed cassia (Chamaecrista rotundifolia) and Russell (1983) are from an intermediate cv. Wynn and Aeschynomene americana cv. Glen. hillslope position. The tops of the hills, at about The main sown grass was setaria (Setaria spha- 70–80 m elevation, were rarely frosted. celata) cvv. Nandi, Kazungula and Narok, with More details of the climate have been given by smaller areas of rhodes grass (Chloris gayana) Stirk (1963) and Cook and Russell (1983) and of cvv. Pioneer and Samford, green panic (Panicum the soils by Beckmann (1957) and Thompson and maximum var. trichoglume) cv. Petrie, Brachiaria Murtha (1960). Detailed analyses of some soil brizantha and Panicum coloratum var. profiles were given by Stace et al. (1968). kabulabula. Some areas, usually on the alluvial soils, were planted to pangola grass (Digitaria eriantha) or sown to kikuyu (Pennisetum The cattle herd and grazing management clandestinum) cv. Whittet. Grazing management of the farm pastures was Until 1990 a breeding herd of about 150 cows at the discretion of the farm manager, and there was run on the station. The breed was initially were 5 managers between 1960–2000. Grazing Hereford but was changed in the early 1970s to during the warmer months tended to be concen- Belmont Red by introducing Afrikander and then trated on the lower areas and this management Belmont Red bulls. After 1990, equivalent strategy was followed far more deliberately over numbers of cattle were maintained by trucking in the last 10 years (1993–2002). Belmont Red weaners from the former CSIRO Narayen Research Station, Mundubbera. Steers were usually retained until turn-off at Experiments 450–500 kg. A small dairy herd of about Experiments varied in size from tens of square 40 Jerseys was maintained for about 14 years. metres to over 20 ha with individual plots ranging from about 1 m2–3 ha. Experiments were Farm sowings usually heavily fertilised, although, if they were laid out on existing farm pastures, they were Farm pastures were usually sown into cultivated rarely fertilised again with the original heavy soil although some legumes were oversown into basal applications (e.g. Jones and Bunch 1995a). existing grass pastures on steeper slopes. Most Some experiments involved planting out of farm sowings were made during the 1960s, lines from breeding programs and many hectares Experiences with farm pastures at Samford 153 of Samford were used in the breeding of Siratro grazing and in lower fertility sites, it did not and later attempts to develop an improved siratro appear to be productive, invariably contributing (Hutton and Beall 1977). Other experiments less than 10% of dry matter. These experiences involved breeding populations of genera such are in accord with research findings at Samford as Stylosanthes, Centrosema, Desmodium, (Jones 1986; 2001) and elsewhere (Wilson et al. Leucaena, Setaria and Digitaria. 1982; Roe and Jones 2000). If seed production Some small-plot experiments involved the problems could be overcome, it or another acces- characterisation of a large range of accessions sion could warrant inclusion as part of a mixture from individual species or genera, such as of species in coastal and subcoastal areas, except Teramnus (Eagles and Pengelly 1996). Other for high fertility or poorly drained sites. Evidence experiments involved assessing different acces- from northern New South Wales suggests that it sions in small plots that were grazed or cut in can improve animal production in such situations common (e.g. Strickland 1978; Strickland and (Mears et al. 1993). Haydock 1978). Small-plot experiments were also used to discover how species reacted to different cutting frequencies and intensities Arachis glabrata (e.g. Jones 1973a; 1974). In other cases, acces- This species has to be planted vegetatively. sions or lines were in individually grazed plots of Although planted in only small areas, usually on about 0.2 ha (e.g. Jones 1979) or in paddocks alluvial or gleyed podsolic soils, it persisted very large enough to record animal production well. One planting of CPI 12121, reported to be (e.g. Jones and Jones 1984; 2003), usually of made in the late 1950s, was still persisting in a 0.5–3.0 ha. Over 40 years, several thousand farm pasture in 2002. Plantings of a wider range accessions would have been sown. of lines in 1970 (Cook et al.