School of Agriculture, Food and Ecosystem Sciences - Theses
Permanent URI for this collection
Search Results
1 - 7 of 7
-
ItemAttributes of pasture influencing the diet of grazing sheep( 2002)This thesis investigated how the composition, productivity and nutritive quality of Phalaris aquatica L. (phalaris) and Trifoliun: subterranean L. (subterranean clover) pasture affected diet selection and intake by sheep in temperate Australia. Field experiments were conducted at the Ginninderra Experiment Station, Canberra, (35 17' S, 149 08' E) and glasshouse experiments at the CSIRO Division of Plant Industry, Black Mountain Laboratories, Canberra, Australia. The effects of different defoliation treatments on the ability of grass and subterranean clover to compete for light were studied in a glasshouse experiment. Regular patterns of phalaris and subterranean clover were planted on a 25 mm grid to create swards with 0% clover, 25% clover, 50% clover, 75% clover and 100% clover by plant number. Swards were either defoliated regularly by clipping to 2, 4, 9 or 15 cm tall or not defoliated and harvested when they reached 2, 4, 9 or 15 cm tall. The distribution of leaf area index within the swards was measured and related to light interception and photosynthesis. Light infiltrated further into the canopy of unclipped swards than clipped swards. Regular clipping created a dense layer of planophile leaves at the top of the canopy, which intercepted most of the incident light, leaving the lower canopy in darkness. Consequently, the lower canopy contributed less to photosynthesis in clipped swards than in unclipped ones. Grass leaves were displayed more prostrate when clipped regularly such that the structure and light interception of grass was similar to clover. In the absence of defoliation, grass leaves were taller than clover, but their erect habit allowed infiltration of light to clover laminae. The relationships between herbage mass, botanical composition, intake and dietary selection by sheep grazing phalaris-subterranean clover pastures were investigated during spring. A range in pasture height (1.2 to 10.2 cm) caused a similar range in herbage mass (370 to 3030 kg DM/ha) and herbage accumulation (-18 to 52 kg/ha/day). A pasture 5 to 6 cm tall, with a leaf area index of 2.0 and an available yield of 1700 kg DM/ha was most productive. The estimated daily pasture intake by sheep ranged from 384 to 1077 g OM. On the shortest pastures, intake was limited by available yield. Intakes increased little after available pasture yield reached 1500kg DM/ha or a pasture height of 5 to 6 cm, but were restricted at pasture heights of 2-3cm or less. There was no evidence for selection in favour of either clover or phalaris, and they were consumed in the same proportion they were present in the pasture. The effect of recent diet on the selection and intake of diet by sheep grazing phalarisclover pastures was investigated. Sheep grazed phalaris, clover or phalaris-clover pasture prior to movement to either a phalaris- or clover-dominant mixture of the two species. The results were complicated by the presence of Vulpia species in the pastures. Where one species dominated the pasture, the diet was predominantly that species. Cases where pasture was a more even mixture of species could not be interpreted successfully, because the alkane concentrations of the pasture species (from which diet composition estimates were made) did not allow the pasture species in the diet to be discriminated with confidence. The daily intake of pasture was affected by previous diet, with animals tending to consume more when moved to a pasture similar to the one they were previously grazing. In an experiment on established phalaris pasture, the diurnal fluctuation in the concentration of water-soluble carbohydrates (WSC) was measured during spring. WSC increased significantly (P = 0.009) from 103 mg/g DM at 0715 hours to 160 mg/g DM at 1300 hours, and did not change further during the next two hours. The concentrations of glucose (17 mg/g DM), fructose (20 mg/g DM), fructan (14 mg/g DM) and "other carbohydrate" (predominantly the carbohydrate moiety of glycosides; 43 mg/g DM) remained relatively constant throughout the daylight period. The increase in concentration of sucrose (33 mg/g DM) was the most significant influence on the increase in WSC (57 mg/g DM). Shading to exclude light prevented the increase in WSC concentration during the day. In a subsequent experiment, a shading treatment was used to create pasture with lower WSC concentration (62 mg/g DM) than an unshaded control (126 mg/g DM). The concentration of the component carbohydrates (i.e. glucose, fructose, sucrose, fructan, "other carbohydrate" and starch) were significantly lower in shaded compared to unshaded pasture. Sheep given free choice between shaded and unshaded pasture exhibited a preference for unshaded pasture over shaded pasture and, on average, 72% of the DM in their diet was unshaded pasture. Whilst no sheep showed preference for the shaded pasture, the proportion of unshaded pasture in their diets varied between 52% and 87% (DM basis). The results are discussed in terms of their implications for grazing management and the development of strategies to improve pasture utilisation and nutritive quality. The role of WSC in diet selection and its importance as a determinant of forage nutritive quality are discussed. The potential benefits of increasing the concentration of WSC in forages by altering agronomic practices or selectively breeding for WSC are discussed.
-
Item
-
ItemThe ecology and physiology of two species of Carduus as weeds of pastures in Victoria( 1977)Slender thistles (Carduus pycnocephalus and C. tenuiflorus) were introduced to Australia about the 1880s. They are now important weeds of pastures in much of southern Australia and are difficult to control with present methods. This study was undertaken to investigate several aspects of the ecology and physiology of the plants with the belief that a knowledge of some of these aspects, particularly of seed germination and seedling establishment, might disclose some "weakness" in the plants' growth which could be exploited to improve control measures. Because of confusion over differences between the two species which occur in Australia the initial step was to evaluate the morphological features which have been used to distinguish between the two species. Although they are very similar morphologically, cytological evidence confirmed that the two species were quite distinct and, in fact, had quite different evolutionary origins. Germination of seeds of slender thistles is controlled by three separate forms of dormancy; these are known as innate, induced and enforced dormancy. Dormancy ensures that the plants will survive in a Mediterranean-type climate and also colonize areas with quite different climates and, most importantly, survive natural catastrophes such as drought, fire, and flood. The germination of slender thistles in the field is confined to a very short period (about 6 weeks) after the autumn break in any year. This is a "weakness" in the plants' survival mechanism because they are vulnerable in that year, at least, to any treatment which can kill seedlings. The herbicide, diquat, was found to kill young seedlings of slender thistles and not affect seedlings of desirable pasture plants associated with the thistles in southern Australia. This treatment is economical and leads not only to a reduction in thistles but an increase of about 30% in pasture production. Several other aspects of the plants' growth were investigated. Slender thistles have early growth characters which give them advantages over more desirable components of pastures. They are more competitive than subterranean clover and ryegrass over a wide range of levels of nutrients, and the traditional approach to pasture improvement in southern Australia of applying superphosphate and sowing subterranean clover will encourage, not suppress, slender thistles. Since grazing animals generally do not eat slender thistles the presence of thistles in pastures at densities commonly occurring in Victoria considerably reduces pasture production.
-
ItemEffects of resistance to prehension and structure of pastures on grazing behaviour and intake of dairy cows( 2000)Pasture intake by dairy cows is affected by plant and sward structural characteristics of the pasture. In the meantime, grazing animals are constrained to gather their food bite by bite removing only a portion of the herbage present at the location which they bite. The extent to which grazing animals overcome the constraints imposed by the plant and structural characteristics of the pasture is the major determinant of herbage intake. The experiments which form the basis of this thesis concentrated on determining the role of sward resistance to prehension, measured in situ, as an integrating sward characteristic that determines foraging decisions of cows and the extent to which they defoliate pasture swards. Under rotational grazing systems, a cow is offered an area of pasture that is often smaller than the area from which the cow harvests its bites (defoliated area, DA). The cow therefore faces changing sward conditions during the process of grazing down into the sward and removing bites at successive lower defoliation planes. The defoliation pattern in grazing down the sward profile and the consequent herbage intake and diet composition, are examined in this thesis. A novel apparatus was designed to measure the BFF in situ at different sward profile heights. In the initial experiment, changes in BFF down the sward profile of six pasture species were examined in order to evaluate the mechanical efficiency of defoliating bites at different depths, in terms of bite weight:BFF ratio. The hypothesis tested was that cows remove 30 - 40% of the sward height at each bite due to a mechanical advantage in terms of BW:BFF. The BFF varied more between defoliation strata than between pasture species. The bite weight and BFF increased with the depth of defoliation. The mechanical efficiency of defoliating bites estimated as the BW:BFF ratio declined slightly with bite depth until a depth of about 30 - 40% of the sward height is reached, when the ratio declined more rapidly. Based on these results and those of Wade (1991), four theoretical defoliation planes (DPI, DP2, DP3 & DP4) were set each at 35% of the pre-grazing sward heights to estimate the total area defoliated by grazing cows under different sward conditions. DP2 is the plane of removal of a second bite after a first bite has removed DPI. Three spring grazing experiments were conducted to explore relationships between pasture allowance and/or sward structure and intake dynamics. In the first experiment, cows were offered a herbage allowance (HA) of 50 kg DM/cow/day either as one block with continuous access for 24 hours, or as six equal break rations opened at intervals during a 24 hour period. In the two subsequent experiments, different sward types were created in order to alter the BFF. In the second experiment swards were created with two different surface heights (USH) and in a 2 x 2 factorial, cows were offered two HA (35 and 70 kgDM/cow/day). In the third experiment, swards with three different tiller densities were created and cows were offered a similar HA of 8 kg DM/cow/3 hours. The defoliation pattern, BFF at 30, 50 and 70% of USH, DM intake, grazing behaviour and the energetics of grazing were measured. The major conclusions derived from these experiments are as follows. The average depth of defoliation (DD) increased with sward height and fell between DP2 and DP4. However, the proportion of area defoliated at each defoliation plane declined down the profile, at rates that varied with HA and tiller density but was unaffected by sward height. At a HA of 70 kg, cows barely reached DP4. The area defoliated at DP4 increased with decreasing herbage allowance and decreasing tiller density. The initial bulk density and post-grazed bulk density declined with USH, but the grazed-stratum bulk density was not significantly affected by USH. Therefore, it was concluded that the volume of canopy defoliated was the major determinant of intake. With increasing HA, the average bite weight (BW) increased, prehension bite rate declined but the overall intake rate increased. The time cost of a bite increased with BW. However, the energy expenditure on prehending a bite did not show a consistent relationship with BW. The BFF increased with sward height and tiller density. However, BFF in the leafy layer of 70% of the sward height was not affected by initial sward height or tiller density. The increase in BFF with initial sward height and tiller density was greater in the lower stemmy layer of 30% sward height. The average bite area (BA) and BW increased with HA. Intake was positively correlated with HA (R = 0.49), HM (R = 0.65) and tiller density (R = 0.51). Multiple regression analysis with herbage intake as the dependent variable indicated that, in addition to HM and HA (R2 = 0.887) , inclusion of the difference in BFF between that at 30% USH and that at 70% USH (BFFdif) as a sward characteristic provided an equation with a substantially better fit (R2 = 0.956). DMI = -3.47 + 1.80 HM + 0.225 HA R2 = 0.887 DMI = -2.73 I + 2.76 HM + 0.732 HA - 0.0416 BFFdif R2 = 0.956 It is concluded that the BFFdif has a significant value in integrating the changes in sward characteristics down the profile and is useful in improving the intake model.
-
Item
-
Item
-
ItemAn analysis of radiata pine-pasture agroforestry systems( 1993)Agroforestry, the integration of forestry and agricultural production, requires an understanding of the interactions between trees, agriculture and the environment. This thesis presents an analysis of the effects of a variety of Radiata Pine-pasture agroforestry treatments on soil water, tree growth and form and pasture production together with a series of economic analyses using the FARMTREE model to simulate the various agroforestry regimes established at Carngham in western Victoria. The Carngham study site consists of a replicated randomized block design of five Radiata Pinepasture treatments. The treatments were: open pasture (no trees), 100 trees/ha (8 m x 12 m), 277 trees/ha-wide-spaced (4 m x 9 m), 277 trees/ha-5 row (5 rows, 4 x 3 m, with 10 row gap) and 1650 trees/ha (no pasture). Results, 11 years after tree establishment, show that soil water content under the various treatments has a cyclical pattern of recharge and discharge with an annual amplitude of approximately 100 mm. Within the 100 trees/ha and 277 trees/ha-5 row and 277 trees/ha-wide-spaced treatments, the soil water content was generally significantly less than under open pasture in the upper 170 cm of the upper profile. Within the 1650 trees/ha treatment, the trees utilize water to a depth of at least 270 cm. Tree form was influenced by tree density. Tree diameter decreased but height increased with tree density with the trees of largest volume produced in the 277 trees/ha-wide-spaced treatment. Variable-lift pruning was routinely applied on an annual basis from tree age 6 years, to minimize the internal knotty-core and to maximise the volume of knot-free timber produced. Pruning was virtually completed (to 6 m) after five or six annual treatments, and significant relationships between DOS (diameter over stubs) and various tree parameters were identified as predictors for determining the volume of the knotty-core. Agricultural production was assessed as net pasture production using rising-plate methodology during the major growth seasons. Trees in single rows, as in the 100 trees/ha and 277 trees/ha-wide-spaced treatments, had only a limited effect on net pasture production compared to the zone within 4.5 m from the trees in the 277 trees/ha-5 row treatment where pasture production was significantly less than in open pasture, or in the zone from 4.5 to 18 m from the tree line where there was a possible shelter benefit. Although pasture production was similar between treatments, animal production decreased with increasing tree density. Economic analyses using the FARMTREE model and progressive data from the Carngham trial showed that simulations of the agroforestry regimes at Carngham were more profitable than grazing alone using real discount rates up to 7%. The 277 trees/ha-wide-spaced regime returned the greatest net present values over the range of discount rates used. The optimum rotation length, assuming a 5% real discount rate, was 26 years. Based on the 277 trees/ha-5 row treatment, shelter benefits for agricultural production could be obtained with a distance between belts of 150 to 200 m and by leaving at least 10% of the trees unpruned. The Carngham trial represents one case study of Radiata Pine-pasture agroforestry for south west Victoria. On this basis, the data presented provides information on patterns and trends likely to be encountered in applying agroforestry to other locations in Victoria. The adoption of agroforestry requires ongoing research trials and demonstrations and the Carngham trial has been well planned and maintained and should be seen as a valuable asset for the advancement of agroforestry in Victoria.