- Title
- The ecology of reproduction in a rare plant - Tetratheca juncea
- Creator
- Driscoll, Colin
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2013
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- Rare species hold a particular fascination for amateur naturalists and professionals alike and a real thrill is experienced when one is encountered in its natural habitat. As a component of overall biodiversity, rare species have received legal protection that requires careful management of these species and their habitat so as to avoid or minimise loss as a consequence of human activity. It is however, inevitable that a continually expanding human populace will result in some loss of diversity, including rare species. The key here is to manage these losses so that they do not lead to extinction and this can only be achieved through a detailed understanding of rare species life-cycle and habitat requirements. This thesis deals with aspects of the reproductive biology of rare plants through the lens of a detailed investigation into Tetratheca juncea, a cryptic, low sub-shrub restricted to a small area of the NSW central and lower north coast of Australia. The currently known distribution of the species divides naturally into two regional populations designated as Central Coast and North Coast. The species propagates through both clonal spread and sexual reproduction from primarily xenogamous hermaphrodite flowers. Clonality poses a problem for determining population density and an arbitrary convention is used defining an individual plant as a clump if separated from the next nearest by 30 cm or more, irrespective of whether it would be a genet or ramet. In 2002, prior to the start of this study in 2003, Tetratheca juncea was reported to consist of no more than 10,000 individual clumps existing in small, highly fragmented subpopulations. A critical weakness in the reproductive cycle of the species was considered to be low levels of pollination inferred from reported low levels of fruit being set, and the fact that no pollinators had been confirmed despite intensive investigation. A literature investigation revealed considerable scope for detailed study to be conducted into phenology, pollination and distribution of this species. The reproductive phenology of Tetratheca juncea was investigated in detail for the majority of one flowering season. Four phenophases: budding, flowering, fruiting and seed release were documented monthly for eight months (June to January) over 12 sites in the Central Coast regional population, covering approximately 30 km north to south. Budding commenced in June and daylight/darkness hours and temperature analysis pointed to photoperiod being the primary initiator. The flowering curve showed a strong positive skew, rising sharply from September, peaking in October and falling slowly, with the suggestion of a second but lower peak in November. Detailed bud length analysis indicated that a second phase of budding in September contributed to the second peak in flowering and extended the overall flowering period. Fruit development commenced in September and seed release in October with both these events being concurrent with budding and flowering past the time that data collection was stopped in January. The flower structure of Tetratheca juncea with poricidal anthers suggested that pollinators would be specialised native bees capable of extracting pollen by buzz-pollination. Six buzz-pollinating native bee species have been confirmed collecting pollen from Tetratheca juncea flowers. These bees are small; visits to flowers are infrequent and even the act of closely observing has been shown to alter pollinator behaviour which makes the task of directly documenting pollinator visits impractical. However, abiotic pollination in the wild appears to be a rare occurrence so the presence of developing fruit can be used as a quantitative surrogate for pollinator activity. Two indices used were Fruit Flower Ratio (FFR) and total fruit per clump, both providing information on different aspects of pollinator activity, flowering and fruiting patterns. Pollinator activity was investigated at both regional population and subpopulation levels. The regional population study was conducted over six years and covered a major portion of the Central Coast regional population. The aim of this study was to determine levels of pollinator activity over time and geographic distribution. Pollinator activity was shown to be highly variable at many levels yet overall FFR values were consistent with those from other studies of hermaphrodite species. There was no indication of pollinator deficit. The subpopulation study was conducted in a group of approximately 500 clumps over six years, mostly overlapping the time span of the regional population study. The subpopulation study provided detail that was not available from the regional population study with its broader scale and opportunistic approach. Again pollinator activity as measured by FFR was found to be within the previously reported range. Mean flowers per clump increased significantly over the study period and mean FFR fell. Analysis of mean fruit per clump showed that pollinator levels remained the same which pointed to falling FFR being a consequence of increased flower numbers. Clump attrition analysis showed a decline in clump numbers of 27% over the six years and habitat data analysis suggested a cause as being increased competition from ground species. Finally, environmental niche modelling was conducted, at a grid size of 100 m, with the aim of understanding the probable distribution and abundance of the species and its broad habitat requirements. Modelling was conducted using the maximum entropy method provided in the program Maxent. Presence only data for the target species are used as input for model development. For various reasons most raw presence data are spatially biased which, if left unchecked, results in a weakened model. A novel method was used to select an unbiased set of Tetratheca juncea presence points that retained the spread of data across the known distribution. Maxent provides several output options and in this case a threshold was chosen that resulted in a binary model showing suitable and unsuitable habitat. The final model was submitted to, and passed, a reasonableness test by comparing niche overlap with a model of a habitat type in which Tetratheca juncea does not occur. Separate models were prepared for the Central Coast and North Coast regional populations. For the Central Coast model, the entire area of interest had been surveyed for the species at varying levels of intensity so the model was run for the whole area. Only a small number of presence data were available for the North Coast area of interest so this model was run in two stages: a buffered common enclosing rectangle around the presence data followed by projection from that model across the remainder of the area of interest. For the Central Coast regional population, the model allowed estimation of the amount of habitat loss since European settlement, habitat fragmentation and total abundance. The North Coast regional population model revealed the intriguing possibility of that population exceeding the better-known Central Coast regional population. This study commenced with Tetratheca juncea considered to be in low numbers in a highly fragmented distribution with low seed production, most probably as a consequence of low numbers of unidentified pollinators. Now, from this study, it can be concluded that the species is widely distributed in two regional populations, albeit over a relatively small geographic range. Actual population numbers could be over two orders of magnitude greater than the 10,000 originally claimed. The species shows few of the attributes of a rare species, being a generalist in its habitat requirements. The species has attributes of being well adapted to its place in a competitive environment. It flowers for longer than pollinators are seasonally available thus maximising pollination opportunities apparently offsetting the disadvantage of having no nectar as an attractant. Specialised buzz-pollinators limit the chances of stigma clogging by pollen from co-flowering species by transferring pollen from their ventral surface. Nectariferous co-flowering species are essential for Tetratheca juncea pollinators and there is a possibility that the presence of both facilitates increased pollinator visits to each species.
- Subject
- Tetratheca juncea; environmental niche model; plant reproductive phenology; pollination; fruit to flower ratio
- Identifier
- http://hdl.handle.net/1959.13/937469
- Identifier
- uon:12565
- Rights
- Copyright 2013 Colin Driscoll
- Language
- eng
- Full Text
- Hits: 1144
- Visitors: 1509
- Downloads: 332
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | ATTACHMENT01 | Abstract | 148 KB | Adobe Acrobat PDF | View Details Download | ||
| View Details Download | ATTACHMENT02 | Thesis | 6 MB | Adobe Acrobat PDF | View Details Download |