Department of Biological Sciences, NUS
Date: 28 Sep 2015, Monday
Venue: DBS Meeting Room (DBS General Office,S3, #05-01)
Supervisor: Asst Prof Chisholm, Ryan Alistair
Presently at least a third of amphibian species globally are listed as at risk from extinction on the IUCN red list of threatened species, and at least a quarter are listed as data-deficient. Considering both the high number of species listed as threatened and the high number of data-deficient species estimated to be threatened, we are unlikely to have the funding or the man-power available to collect adequate baseline data, or apply conservation measures for all species. This thesis is thus aimed at addressing gaps in our current knowledge by utilizing mathematical modeling tools combined with modern technological tools for optimal monitoring and survey design. These methods can help to allocate scarce resources efficiently. I aim to address four main areas in this thesis: (1) Is it possible to use data-rich species to inform what type of training conditions are required in Species Distribution Models to create reliable predictions for data-poor species? Specifically does the size of the region the model is trained on vary according to genus and life-history traits? (2) Can we design more efficient monitoring programs and surveys for biodiversity and single species and what type of pre-requisite data is needed? (3) How many individuals are detected when performing a survey and how is this influenced by the skills of the surveyor and the traits of the species? (4) Is there a need to have acoustic monitoring stations distributed vertically in the tree and can we improve on the current classification algorithms for acoustic data?
All are welcome