Using new computational tools to investigate the responses of cotton plants (Gossypium hirsutum L.) to defoliation

Thornby, David (2004). Using new computational tools to investigate the responses of cotton plants (Gossypium hirsutum L.) to defoliation PhD Thesis, School of Land, Crop and Food Sciences, The University of Queensland.

       
Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
Thornby_phd_totalthesis.pdf Thornby_phd_totalthesis.pdf application/pdf 1.60MB 1
dt_thesis_pkg.zip dt_thesis_pkg.zip Click to show the corresponding preview/stream application/x-zip 5.19MB 0
Author Thornby, David
Thesis Title Using new computational tools to investigate the responses of cotton plants (Gossypium hirsutum L.) to defoliation
School, Centre or Institute School of Land, Crop and Food Sciences
Institution The University of Queensland
Publication date 2004
Thesis type PhD Thesis
Supervisor Room, P.
Adkins, S. W
Hanan, J. S.
Total pages 251
Collection year 2004
Language eng
Subjects 300000 Agricultural, Veterinary and Environmental Sciences
L
270402 Plant Physiology
620107 Cotton
Abstract/Summary Plant responses to damage, including defoliation, are complex. Plants have been shown to have a marked ability to compensate for damage, including by producing new or larger fruiting structures and by regrowing vegetative components following damage. While compensation for the loss of reproductive components has been widely investigated, compensation for the loss of vegetative components has been much less thoroughly studied. Similarly, biomass has been used as the index for studies of compensation in plants, while spatial aspects of compensation have been ignored. The spatial aspects of compensation include the topological position of lost and compensating structures, which is an indicator of resource allocation strategies in the plant following damage. Due to the complexity of interactions between structural and non-structural aspects of plant growth, particularly in response to damage, a computational approach to the problem is likely to provide advantages over more traditional approaches. In computational plant science, a simulation step may be added to the established scientific paradigm of hypothesis-experiment, so that it becomes hypothesis-simulation-experiment. The simulation step formalises what is currently known and allows for the generation of potentially more relevant and more testable hypotheses than ad hoc methods. Cotton (Gossypium hirsutum L.) is an important crop in Australia, and a significant part of the agricultural economy. Cotton’s growth is constrained by a number of environmental factors, including the activities of pests. Cotton’s morphological responses to damage are particularly complex. Accordingly, cotton was chosen as a suitable model crop for investigating morphological responses to damage with computational tools. L-system models of cotton morphogenesis were produced by linking hypotheses about physiology, initially from the literature and later from data gathered during the research, with the structural outputs of physiology. The resulting simulations were used to generate morphological hypotheses that could be tested with real plant experiments. During the experiments, morphological data were gathered with a sonic digitiser, allowing comparison of the architectures of cotton plants at the level of each individual component. Because the measuring technique was non-destructive, plants could be measured many times, creating a set of data representing dynamic plant growth. Relationships from the data for several experiments on cotton's responses to defoliation of the main stem and branches of seedlings and older plants were used to adjust the initial model, rejecting some physiological hypotheses and confirming others. The result is a model that represents aspects of the interaction between cotton's carbon allocation, topology, and signalling behaviour following defoliation. Initial experiments found that apical development rates are not limited by photosynthate supply, at least at the levels of defoliation applied here. Unperturbed, obligate apical development was shown by subsequent versions of the model to be a significant contributor to cotton's ability to compensate for early season defoliation. Experiments on branch defoliation illustrate that while the vegetative branches respond in ways that are similar to those of the main stem, the sympodial branches behave in qualitatively different ways. Sympodial branch development was found to be significantly perturbed by the loss of the subtending main stem leaf, somewhat perturbed by the loss of significant leaf area within the branch, and very seriously perturbed by the removal of both the main stem leaf and branch leaves. The ability of a branch to produce mature bolls was likewise severely affected by defoliation, either of the branch leaves or of the subtending main stem leaf. However, no effect was noted where defoliation was applied to nearby branches or main stem leaves. This supports the theory that cotton branches are independent physiological units with regards to photosynthate. Branch independence interacts with cotton's propensity to produce many more fruiting structures than it can mature as a compensation mechanism for defoliation and subsequent fruit loss. This research demonstrates the utility of the hypothesis-simulation-experiment approach to biological investigation. This approach provides a way to use qualitative and quantitative physiological knowledge to produce readily tested morphological hypotheses for use in real plant experiments. The results of such experiments are demonstrated to be not only empirical evidence of a response, but evidence in support of or contradicting proposed physiological mechanisms for the response. Thus, the hypothesis-simulation-experiment approach as used in this thesis represents a method for generating and testing more meaningful hypotheses than are generally produced using ad hoc hypothesis generation techniques.

 
Citation counts: Google Scholar Search Google Scholar
Access Statistics: 127 Abstract Views, 1 File Downloads  -  Detailed Statistics
Created: Fri, 08 May 2009, 08:20:32 EST by Noela Stallard on behalf of Library - Information Access Service