School of Agriculture, Food and Ecosystem Sciences - Theses
Permanent URI for this collection
Search Results
1 - 10 of 45
-
ItemImproving species distribution models using extreme value theory and climate dataset ensembles( 2020)The development of climate datasets at fine spatial and temporal scales has commonly been driven by the need to better understand vegetation distributions and ecological systems. While a wide range of global, national and regional climate datasets have been developed over the last two decades, they are rarely compared directly in the ecological literature. This thesis evaluates a range of climate interpolation techniques and investigates how the spatial and temporal characteristics of climate datasets may be utilised to improve the predictive performance of plant species distribution models (SDM). A series of spline-based and geostatistical methods for interpolating temperature variables are first compared across Victoria, southeast Australia. Secondary predictors (thermal remote sensing data and local topographic indices) which indirectly capture mesoscale microclimate and cold air drainage regimes were found to improve monthly mean minimum temperature interpolations by up to 39%. Thermal remote sensing data only reduced root mean square error (RMSE) by up to 6% for maximum temperature across Victoria and was most effective during the summer months. The interpolation methods used in southeast Australia were subsequently transferred to the Royal Himalayan Kingdom of Bhutan to validate their effectiveness in a novel climate. In Bhutan, the predictive performance of minimum temperature interpolations was also improved considerably (up to 23% reduction in RMSE) when using thermal remote sensing data and local topographic indices as spatial covariates. Thermal remote sensing data also reduced the RMSE for maximum temperature interpolations by up to 16% in Bhutan. Interannual variability of climate extremes were used to evaluate how the temporal characteristics of climate may be used to improve the predictive performance of SDMs. Generalised Extreme Value (GEV) distributions were fitted to monthly climate data to generate variables which account for the skewed distribution of extremes. Models incorporating interannual variability (drawn from a range of expected return intervals) improved predictive performance compared to models using seasonal extremes only for 28 of 37 species assessed. Iteratively fitting models using alternate expected return intervals typically acted on the leading and trailing edges of current distributions, indicating that such methods may be useful for model calibration and characterising climate-driven source-sink population dynamics. The impact of spatial disparities in climate on the predictive performance of plant SDMs was evaluated using three distinct datasets developed for Victoria as part of this research, in addition to two global datasets (WorldClim v1 and v2). Individual models were compared against one another and as ensembles to explore the potential for alternate predictions to complement one another. The Victorian datasets demonstrated a significant improvement over the original WorldClim dataset (up to 17.3% mean increase in D2) and trended towards an improvement relative to WorldClim v2; however, no significant differences were found when comparing the alternate Victorian datasets. Multi-model ensembles achieved a mean increase of up to 13.8% and 29.2% in D2 relative to individual models when using regional and global datasets, respectively. Ensembles provide a pragmatic method to improve the predictive performance of SDMs and allow a trade-off between the uncertainties and potential biases embedded in competing climate datasets.
-
ItemMaking the connection between history, agricultural diversity and place: the story of Victorian apples( 2016)Apple growing practices are embedded in a productivist mentality aiming for ever higher efficiency and productivity. And while the climate change impacts are to a large extent known, there is little attention given to the coupling of the social and the ecological effects. I use apple growing as a case study to explore the relationship between place, biodiversity and rural change in Victoria. My research is based on historical research; including an analysis of the Museum Victoria’s collection of wax apple models, and in-depth interviews with orchardists. By drawing on environmental history, social-ecological systems thinking and Bourdieu's theory of practice, I highlight the importance of a systems perspective and inform it by emphasis on the critical role of underlying power structures and individual dispositions, or the habitus, of the growers. These dispositions have been shaped and internalised by the growers’ histories and their physical surroundings. Orchardists have been able to respond to intensifying production requirements by utilizing technologies and scientific nous to keep up with the continuous aim for efficiency. Growers are caught up in a self-reinforcing cycle of satisfying the demand for perfect apples by adopting expensive techno-scientific approaches to enable ever more intensive production. The symbolic violence and amplified biophysical pressure orchardists experience has driven many to despair; resulting in a significant decline in small scale apple growing businesses over the last decade. I offer some suggestions for government policy and support measures and argue that any services or support programs need to be tailored to the appropriate level and need of each orchard business and the individuals who are involved. My analysis shows that those growers, who engage more closely with their biophysical place as well as their history and identity as apple growers in that place are (re-)creating another version of what it means to be an apple grower. In some cases this is resulting in resistance to the vortex of agricultural productivism that has been the basis of their existence for many generations.
-
ItemEffects of adding nutrients on soil chemistry and tree growth in native Eucalyptus forests of south-eastern Australia( 2007)The decreasing area available for timber extraction in south-eastern Australia, due largely to social pressure to reserve greater areas of forest, has led to the consideration of fertiliser-application to increase wood output from the remaining available forest. Potentially deleterious effects of fertilising on water quality must be assessed before implementation on a wide scale. This is in accordance with relevant forest management policies. This study examined the effects of applying fertilisers containing nitrogen and phosphorus, on soil and soil-water chemistry in two pole-sized stands of mixed Eucalyptus spp in the Wombat Forest, in the Midlands Forest Management Area, Victoria, Australia. The findings are synthesised and discussed in relation to management of regenerating mixed-eucalypt forests in south-eastern Australia. Fertiliser treatments were none (R); 400 kg N ha-1 as ammonium-sulphate (N); or 400 kg ha-1 plus 202 kg P ha-1 as triple superphosphate coated with 10% sulphur (NP). It was calculated that incidental additions of S were 1371 kg ha -1 (N treatments), and 1696 kg ha-1 (NP treatments). It was expected that P would be principally adsorbed on soil surfaces; N immobilised in the soil organic pool and that metallic cations would enter the soil solution to varying degrees. Fertiliser-addition increased both plot-basal-area (BA) growth and the rate of stand self-thinning. In 3.8 years, BA in reference (R) plots at two sites increased by 7.3% and 23.4%. Where N alone was added, BA increased by 14.2% and 27.1%, while in NP plots BA increased by 17.1% and 42.7% respectively. Mortality was 9% in untreated plots compared to 14% in NP plots. Estimated increases in biomass growth equated to additional above-ground nutrient accumulation of 0.4 to 1.5 kg ha-1 of P, and 5.5 to 20.8 kg ha-1 of N. This represented only 0.2 to 0.7% of added P, and 1.4 to 5.2% of added N. Soil solution was extracted from 10 and 50 cm with porous-ceramic-cup tension-lysimeters (-0.6 kPa). Concentrations of P and N were low both before and after adding fertiliser. Across all treatments the maximum median PO43- concentration in soil-water at 50 cm was 0.12 ppm (mean 0.28 ppm). Typically PO43- concentrations were not higher than 0.03 ppm. The 400 kg ha-1 of added N was rapidly immobilised in the soil organic pool. The greatest mean NH4' concentration from a single sampling occasion was 1.1 ppm. The mean NO3 concentration at 50 cm was never higher than 0.26 ppm. After adding N in fertiliser the proportion of NO3- relative to NH4* in soil-water increased and was correlated with decreasing soil-water pH. Less than 1% of added P and N was recovered from soil solution at 50 cm. The largest pool of added P recovered was PO43- adsorbed to soil between 0 and 20 cm, due to the soil adsorption capacity being well in excess of the applied 202 kg P ha-1. Phosphate desorption using sequential extractions with a mild acid extractant (0.3M NH4F, 0.1M HCI) recovered between 25% and 116% of added P. Differences were attributed to both the amount of P added and the effect of time since treatment at different sites. Soil disturbance during sampler installation was found to be more likely to raise soil-water P concentrations at 50 cm than would adding up to 202 kg P ha-1. Among the ions in solution. SO42- and CI' were the dominant anions while Cat+ dominated the cation chemistry. In untreated forest 5042- in soil-water ranged from 7.7 to 16.0 ppm at 10 cm and 7.9 to 12.2 ppm at 50 cm. In fertilised plots up to 100.5 ppm SO42 was measured in soil-water at 50 cm depth. In the N treatment at 50 cm, SO42- in soil-water accounted for 9.4 % of applied S. compared to 14.0 % in NP. In untreated forest, soil-water Cl- and SO42- accounted for over 98% of the total soil-water anions, in roughly equal proportions at 10 cm, and CI- slightly higher at 50 cm. Following fertiliser-application soil-water pH at 10 cm fell from 6.3 in R to as low as 4.81 (N) and 4.45 (NP). At 50 cm pH never dropped below 6 and there were no visible departures from reference concentrations. Relative activities of K+ and Mg2+ in solution increased with decreasing pH, indicating increased leaching potential. Sulphate in soil-water increased total anion charge further in NP than in N. Total charge (cmolc L-1) for cations followed anions. A slight deficit in anion charge was likely due to the unquantified contribution of organic anions. These results confirm that despite the quantity of fertilisers added in this trial being double likely operational quantities, the forest and associated soils had the capacity to retain these nutrients through a variety of processes. The study validates the environmental sustainability of proposed intensive management practices including fertiliser-application in this forest type. It also emphasises the importance of understanding fundamental forest nutrient cycling processes when aiming to carry out intensive forest management practices in an environmentally sensitive manner.
-
Item
-
Item
-
ItemIdentification of boron tolerance in Brassica rapa( 2006)There has been increasing interest in developing canola quality B. juncea for low rainfall areas across Australia over the past two decades. However, B. juncea genotypes are susceptible to high levels of boron in Western Victorian soils. An understanding of the genetics and the molecular basis of boron tolerance may enable fast and accurate tolerance selection and lead to improved boron tolerance. Being an allotetraploid species, B. juncea is difficult to understand at the genetic level because of chromosomal duplication and the potential presence of multiple copies of the loci of interest. Therefore, once the tolerance genes or chromosomal loci governing tolerance are identified in the diploid progenitor genomes, B. rapa and B. nigra, boron tolerant B. juncea lines may be resynthesized. Thus, as an initial step in this process, this thesis aimed to understand the physiological, genomic and molecular mechanisms involved in boron tolerance in B. rapa. Initially, B. rapa genotypes were screened for tolerance to boron toxicity using hydroponic and soil assays. On the basis of primary root length, severity of leaf toxicity symptoms, dry matter accumulation and shoot boron uptake, the B. rapa genotypes WWY Sarson and Local were identified as the most tolerant and the B. rapa genotypes Shillong and Kaga the most susceptible to toxic boron concentrations (1000 ?M B in hydroponic assay; 54 mg B kg-1 soil in soil assay). The main mechanism of tolerance to boron toxicity in B. rapa involved reduced net boron uptake by roots, with some boron accumulation in the tap roots and partial exclusion of boron from shoots. Furthermore, boron uptake was much lower in the WWY Sarson and Local genotypes than in the Shillong genotype, despite higher rates of transpiration. This implied that an active boron efflux mechanism may be operating in the tolerant genotypes. The inheritance pattern of tolerance to boron toxicity in B. rapa genotype, WWY Sarson best fitted a Mendelian model of two major dominant and epistatic genes. A B. rapa linkage map was constructed from an intraspecific F2 population (WWY Sarson X Shillong) with ISSR, RAPD, SRAP and SSR marker loci. The linkage map spanned a total length of 874.1 cM and contained 12 linkage groups. Chisquare analysis (P < 0.05) revealed 25 dominant markers that showed segregation distortion in the F2 progeny. QTL analysis using composite interval analysis identified three significant peaks on LG2 and LG8 that were associated with primary root length and which accounted for 17% of the trait variation. Differential transcript analysis of SRAP markers following exposure to a toxic boron concentration identified up-regulation of me4+em2570bp, me2+em2650bp, me2+em1 1600bp, me2+em1800bp and me4+em2500bp genes in Shillong and Kaga and down-regulation of me2+em2650bp, me2+em1 1600bp, me2+em1800bp and me1+em21200bp genes in WWY Sarson and Local. Of these, a UDP-glycosyltransferase gene (sharing 80% similarity to the Arabidopsis thaliana homolog) was highly transcribed only in the sensitive genotype, Shillong, and may be involved in excessive boron cross-linking to the glycosyl groups present in the cell walls and/or membranes eventually causing the observed reductions in shoot and root growth.
-
ItemEpidemiological and physiological studies of the effects of peach rosette and decline disease on the peach, prunus persicae L. Batsch( 1975)The incidence in the field of the disease peach rosette and decline (PRD), which is of considerable economic importance in the Goulburn Valley, causing fruit loss and tree death, was shown to increase from 0.9 to 91.3% in an orchard of cv. Golden Queen in 10 years. Similar results were found with the cv. Pullars Cling, in which infection increased from 1.5 to 29.7% over five years. The pattern of spread was mainly from infected trees to contiguous uninfected trees. This is consistent with the view that the main causal agent, prune dwarf virus (PDV), is transmitted only via the transfer of infected pollen : a previous finding in cherries which was confirmed in peaches. Prunus necrotic ringspot virus (PNRV) is the other virus always present in the field in PRD-infected trees. The mode of spread of PNRV is also by pollen. Within the tree, PDV moved erratically from the first infected limb, via phloem but not xylem, into the other limbs well in advance of the appearance of symptoms. Three months after flowering, PDV was detected in 65% of main limbs adjacent to the first infected limbs but in only 30% of limbs more remotely positioned on the tree. However, removing infected limbs within four weeks of flowering, when the initial infection was presumed to have occurred, did not prevent the movement of PDV into the rest of the tree. Laterals from peach trees infected with PRD were tested for the presence of PDV, using woody virus indicators (cvs Golden Queen, Italian Prune and Elberta seedling). Golden Queen was found to be a more reliable indicator for detecting PDV than Italian prune, as the presence of PNRV with PDV killed 71% of the Italian prune buds compared to only 34% of the Golden Queen buds. Golden Queen also developed more obvious foliage symptoms of PDV infection than Elberta seedlings. The probability of failing to detect PDV in infected field trees, using all three indicator plants, was higher in the first year of infection. The rate of spread of PRD was reduced in the orchard by preventing infected trees from flowering, either by removing obviously infected trees or by deblossoming. Removing infected trees resulted in a three-fold reduction in the spread of the disease in two seasons. Removing the flowers from infected trees before pollination reduced the spread of the disease by about half. This, only partial, control of the spread of PRD by tree removal or deblossoming was attributed to the presence of up to 14.3% of trees without symptoms being latently infected with PDV. It was observed that deliberate infection with PDV by pollen also resulted in a slow expression of the symptoms of PRD. The effects of PRD on the growth of young peach trees was obvious in the first three months of growth. There were considerable varietal differences in the severity of this effect. Those varieties based on cvs. Golden Queen or Levis Cling were more severely affected than the variety Elberta. The results from shoot elongation measurements agreed with those obtained from conventional growth analysis methods. These latter experiments showed that, after three months, the dry weight and leaf area of infected Golden Queen plants were reduced by 94%. The fruit yield from mature PRD-free trees was three times that of trees infected for the first season, even though symptoms were apparent only on one limb; and six times that from chronically affected trees infected for two seasons. The effect of virus infection on the photosynthetic ability of single, attached peach leaves was studied under laboratory conditions using infra red gas analysis. The constants derived from the equations describing the relationship between net photosynthesis (Pn) and both irradiance and CO2 concentration were used to analyse the effects of infection by PRD on photosynthetic characteristics of the leaf. The asymptotic value of Pn (Pmax) in young leaves was reduced 15% by PRD-infection, mainly through an increase in the "residual resistance" to 002 diffusion and a decrease of 23% in the parameter indicating photochemical efficiency. There was also evidence that the gas phase resistance was higher in infected leaves at low levels of irradiance. Dark respiration was 51% higher in infected leaves, but this difference was not significant. PRD did not reduce Pn in 60-day-old leaves, normal leaf senescence having a predominant and greater effect. It was concluded that PRD infection had its large effects on growth via a reduction in leaf area; the effects on the photosynthetic capacity per unit leaf area being minor. An effect of PRD infection on the translocation of 14C-assimilatesout of leaves was also observed. Infected leaves retained twice the assimilates than did uninfected leaves. It is concluded that the most promising methods of control of PRD include removal of infected trees, deblossoming suspected infected trees until diagnosis is confirmed, use of virus-tested plants, the gradual destruction of infected orchards and protecting young, healthy orchards from infection either by barrier crops or deblossoming the young plants until they reach an economic bearing age.
-
ItemGrazing ecology and high producing dairy cows( 2005)This body of work (82 papers in scientific journals and 2 books) encompasses two broad areas of work. They are 1) growth, nutritive value and management of pastures grazed by dairy cows (40 publications), and 2) supplements for grazing dairy cows, with a particular focus on responses associated with supplement use and digestion in the rumen (44 publications). These two areas of research are inextricably linked and, taken together, have been termed `Grazing Ecology'. Of the 84 publications included, the candidate was the senior or sole author of 67% of them. The chronological development of the work reported includes research on stocking rates reported in the early 1980's through to the development of Diet Check, a decision support tool incorporating much of the information generated during the previous two decades, in the early 2000's. The publications cover aspects of grazing management to optimise growth, persistence and nutritive value of irrigated annual and perennial pastures for dairy cows. Most of this research has incorporated some aspect of stocking rate, whether it be stocking rate per se in long term experiments or frequency and/or intensity of defoliation in shorter term experiments. The aim was to establish optimum grazing strategies that best effected the compromise of maximum intake of pasture of high nutritive value while satisfying the requirements for maintenance of pasture growth and persistence of a balance of desirable pasture species. The research allowed the definition of the intake and nutritive characteristics of pasture grazed by lactating dairy cows under a range of management conditions. At the same time, strategies to effectively feed supplements were investigated. When more than one feed is offered to dairy cows, associative effects play an important role in the eventual responses achieved. Balance of nutrients, particularly in the rumen, and substitution of supplement for pasture in the diet of grazing dairy cows, were the main aspects of the associative effect between feeds considered in the research reported here. Substitution can have a huge effect on the responses obtained from supplements, and the type of supplement, by influencing the balance of nutrients ingested into the rumen, affects the composition of the milk produced. Finally, some attempt has been made to draw much of the information on pasture management and supplementation of grazing dairy cows together for use by dairy farmers and their advisers, and to define gaps in knowledge. This has been done by reviewing the scientific literature, and by the use of modelling to provide simple tools for tactical decision making. Although the research was undertaken in northern Victoria, many of the results apply equally in other areas of the world where pasture constitutes a major proportion of the diet of dairy cows. Victoria currently produces more than 60% of Australia's milk, with northern Victoria producing more than 40% of that. The development of dairying in Victoria mirrors much of the progress of the research reported in this collection of scientific publications. Before 1982, dairy farming was almost totally based on grazed pastures and the use of pasture supplements (hay and silage). A severe drought occurred in 1982, which prompted a serious consideration of the use of supplements for lactating cows grazing pasture. Today, dairy systems in Victoria vary to a huge extent, with the energy provided by pasture ranging from 0 to 100%. Over this period, average milk production has increased, from about 3000L/cow per lactation to more than 5000L/cow. With a fine line separating profit and loss in dairy businesses that basically depend on the price received for manufactured products on overseas markets, both grazed pasture and supplements need to be used optimally. The challenge has been to provide information and tools to allow dairy farmers to achieve this objective. I believe that my research, particularly in relation to pasture intake, substitution and associative effects, has been instrumental in allowing pasture-based dairy farmers to continue to remain viable in Australia, and that many of the principles developed apply wherever pasture constitutes a significant proportion of a cow's diet.
-
ItemPeach tree water use( 2004)Peach tree water use (TWU) was measured and simulated in order to improve both current understanding of water use and to devise a practical irrigation scheduling method for orchards in northern Victoria. TWU was appraised in terms of a basal crop coefficient (Kcb) and reference crop evapotranspiration (ETo). Weighting of ETo by measures of vegetative ground cover and solar radiation interception were explored as techniques to account for changes in TWU associated with tree size, row orientation and tree training method. Sap flow (SF), using the compensation heat pulse technique, was evaluated as a method to measure TWU. Field experiments explored the variation in sap velocity (SV) in the trunk of peach trees and the accuracy of the technique to measure TWU. Sap velocity varied radially across the sapwood so a sap velocity profile was used to calculate sap flow (SF) from measurements made at specific depths in the sapwood. Analysis of the variation in SF around the circumference of a tree revealed the need for at least four sensors installed at different positions in the trunk. Calibration using the cut-tree method established the relationship: TWU = 1.44 (� 0.04) SF. A short-term experiment was conducted to determine effects of reducing tree size on TWU. Tree size was progressively reduced by de-branching an individual isolated tree over a 15-day period. Measures of TWU by sap flow were compared with estimates derived from ETo, and either canopy cover (CC; estimated from the horizontal extent of the canopy) or the area of shade cast by the tree on the soil surface (ASH). ASH was estimated prior to each de-branching event using a combination of photographs of the tree taken from the direction of the sun together with measures of fractional radiation interception in the area of shade cast by the tree. TWU and ETo averaged 42.1 litre/d and 4.7 mm/d, respectively, in the 6-day period prior to de-branching. CC and effective canopy cover (ECC; estimated as ASH measured at solar noon) were 7.8 and 5.8 m2 respectively, in that period. Five de-branching events reduced TWU, CC and ECC by > 95 %. To account for the daytime variation in ASH, effective area of shade (EAS) was calculated from estimates of ASH at solar noon and 3 h each side of solar noon. Kcb, the basal crop coefficient defined by Allen et al. (1998), was related to EAS by Kcb = 1.12 EAS. The effects of row orientation on TWU were investigated by a combination of field and modelling studies on isolated trees and hypothetical hedgerow orchards. TWU was measured by sap flow in six isolated field-grown trees pruned in a hedge shape and orientated north-south (NS) and east-west (EW). TWU was simulated by weighting ETo by the area of shade cast by the tree on the horizontal soil surface (ASH). The model effectively weighted ETo (`Big Leaf') by the fraction of direct beam radiation intercepted by tree foliage in orchards. ASH was estimated using a radiation interception model. Maximum rates of TWU in NS trees were attained in the morning when ETo was approximately half its maximum, and these rates were maintained for much of the day, whereas TWU for EW trees increased linearly with increase in ETo. Leaf area density (p) was estimated as 1 m2/m3 by comparison of observed and simulated TWU and was subsequently used in simulation of ASH in hedgerow orchards. The maximum rate of simulated TWU (TWU') occurred approximately 3 h each side of solar noon in a NS orientated hedgerow but declined in the middle of the day. Shaded leaf transpiration was suggested as a possible contributor to TWU that was not considered in the radiation interception model. The study demonstrated that EAS-weighted ET0 provided a reliable estimate of daily TWU irrespective of row orientation, leaf area density and hedgerow width. TWU and ASH were compared in Tatura trellis (TT) and central leader (CL) orchards. Field observations indicated that TWU in TT orchards attained a maximum near solar noon (matching ET0) whereas in CL it reached a maximum in the morning then remained constant during the middle of the day. Measured ASH revealed a substantial depression at solar noon in CL compared with TT. ECC was 2.3 and 1.7 m2 in the TT and CL trees, respectively. By comparison, EAS was 14 and 30 % greater than ECC for TT and CL trees. These differences between ECC and EAS were even greater in simulated hedgerows taller than used in the field experiment, leading to considerable variation in the relationship between simulated ECC-weighted ETo and TWU. On the other hand, the relationship established between simulated EAS-weighted ET0 and TWU for hypothetical TT trees (TWU = 1.17 EAS ETo) was the same as that established for isolated trees. The results of this thesis suggest that TWU of well-watered trees is slightly above the unit rate of water use described by ET0 (Allen et al. 1998) over the area of shade cast by the tree on the soil surface. Assuming negligible understorey water use, irrigation amount can be simply estimated from 1.12 EAS ETo where EAS accounts for daytime changes in ASH associated tree size, row orientation, training system, and leaf area density. Given this, irrigation requirement can be calculated from ETo and routine measures of orchard EAS. This is a simpler procedure than current FAO recommendations based on ET0 that require a look-up table of Kcb values for different growing periods and further adjustments for vegetation cover. In contrast, the EAS-weighted ETo proposed here provides an objective estimate of TWU for which EAS can be readily established by routine measurements of the fraction of shade in an orchard by irrigation consultants, extension officers, or growers.
-
Item
- 1 (current)
- 2
- 3
- 4
- 5