Use this URL to cite or link to this record in EThOS:
Title: Diospyros in west Africa : morphology, molecules and climate
Author: Bayor, Hypolite
ISNI:       0000 0004 2726 9253
Awarding Body: University of Reading
Current Institution: University of Reading
Date of Award: 2012
Availability of Full Text:
Access from EThOS:
Diospyros is a pantropical genus occurring in the lower layers oftropical moist forests. As such the persistence of the forest may be critical to the survival of some species. Twenty-two species of Diospyros occur in the Upper Guinean biodiversity hotspot, seven of which are endemic. The effectiveness of vegetative characters used with computer-aided keys, DNA barcoding, phylogeny and species distribution modelling as tools to study the biodiversity of this area was investigated. A computer-aided multi-access key developed indicated that vegetative characters might be able to identify some species but more testing is required. The effectiveness of the proposed barcodes for plants (rbeL and matK) and psbA-trnH were tested. Bayesian trees showed that single regions produced unresolved trees but using the three regions combined could assign 26 accessions to species. Using TaxonDNA, matK and psbA-trnH were almost equally effective at 84.6% and 84.0% respectively in assigning field collected accessions to species however, rbeL identified only 37.4%. For a combined data set of Genbank and field data, proportion of accessions correctly assign to species was lower. TaxonDNA assigned 59.6%,49.3% and 75.1% accessions of rbeL, matK, and psbA-trnH respectively to species. Phylogenetic analysis using rbeL and matK shows that West African species may belong to at least four different lineages and these are scattered over the entire worldwide Diospyros phylogeny. Sister species mayor may not have overlapping ranges indicating that dispersal might contribute to speciation in Diospyros. The effect of climate change on five ofthe endemic species was investigated. Although reduction in predicted area for three species was observed, the area occupied by one was not affected while one species was predicted to have potential to expand its range which suggests genetic homogenization as a possible outcome. The availability and quality of data for species distribution modelling offive species was also investigated. Data were insufficient for two species not modelled and gaps in sampling were also evident. The effect of biased geographic sampling was to inflate AUC values of distribution models. Overall, it is clear that technologies such as computer-aided keys, DNA barcoding, phylogenetic analysis and climate envelope modelling help to study and understand diversity in this Western African biodiversity hotspot. The identification tools can be challenging to develop but will allow better surveys and in turn help to fill the data gaps revealed in the climate envelope modelling studies.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available