http://theses.ncl.ac.uk/jspui/handle/10443/77 2026-04-13T05:14:01Z 2026-04-13T05:14:01Z Girardello, Marco http://theses.ncl.ac.uk/jspui/handle/10443/6393 2025-03-04T15:44:14Z 2009-01-01T00:00:00Z

Title: The application of tree-based methods to species distribution modelling Authors: Girardello, Marco Abstract: Species distribution models are used increasingly in both applied and theoretical research to predict how species are distributed and to understand attributes of species’ environmental requirements. This thesis aims to explore the application of tree-based methods to species distribution modelling. Although these methods have been widely used in other fields of science they have received relatively little exposure in Biogeography and Conservation Biology. The techniques applied include CART, Bagging, Random Forests and Boosted Regression Trees. These were used with four different biodiversity databases to answer different a variety of research questions aimed at: (i) understanding how landscape structure and climate affect species distributions (ii) predicting the potential impacts of climate change on species distributions (iii) to identify areas important for biodiversity conservation. Additionally, the performance of each method was compared with the aim (iv) of making suggestions for the optimal models which should be used by future researchers. In chapter 2 Boosted Regression Trees were used to quantify the importance of wetland size and weather patterns for waterbird distribution in Britain. As well as revealing the importance of wetland size for waterbirds, , the models proved to be reasonably robust when validated. In chapter 3 this basic form of modelling was expanded, using a database containing amphibian occurrence records for Italy. Random Forests was used to quantify species-climate relationship and to predict amphibian distribution in relation to current and future climate conditions. The results revealed how amphibian distribution is largely controlled by temperature-related variables and highlighted a negative response to future climate changes in most species. In chapter 4 Bagging was used to identify areas important for biodiversity conservation. Specifically, Bagging was used predict the distribution of 232 species of Butterflies in Italy. The predicted surfaces were then used in combination with a species multispecies prioritization tool in order to identify important areas for butterfly conservation. The results iii showed that the most areas important for butterfly are located within the Alps, the mountains of central Italy and the island of Sardinia. Finally, in Chapter 5, the predictive accuracy of four modelling techniques based classification trees was compared. This was done using large scale bird distribution data from Italian Common Bird Census. The results showed that Random Forests and Boosted Regression Trees were the best performing techniques and that model performance was highly influenced by species ecological characteristics as well as by the modelling method. The results of this thesis have shown how tree-based modelling methods can be used for exploring and testing hypotheses about the factors that are important in determining species distribution and making predictions of species distribution for use in conservation contexts. The methods used represent a useful way to visualize and understand relationships between environmental parameters and species distributions and to predict species distributions with high accuracy. Whilst it is true that some tree-based methods can be used instead of statistical modelling techniques others expand the analytical opportunities by enabling analyses that are impossible or very difficult with statistical methods. Hopefully this thesis will serve a source of inspiration for ecologists willing to move away from statistical inference and the P-value dogma and concentrate on understanding the data, and using alternative techniques to predict species distribution with high accuracy. Description: PhD Thesis

2009-01-01T00:00:00Z Mordue, Simone Michelle http://theses.ncl.ac.uk/jspui/handle/10443/4550 2019-11-06T14:58:40Z 2019-01-01T00:00:00Z

Title: How does the spatial and social dynamics of the Natterer's bat Myotis nattereri affect disease transmission and conservation? Authors: Mordue, Simone Michelle Abstract: Natterer’s bats (Myotis nattereri) are typical of many Bat species in that they participate in a variety of distinct seasonal communities and behaviours. In summer adult females are thought to be largely philopatric to their natal community/landscape where they rear their young and form largely matrilineal communities. Bat foraging behaviour and social participation is largely unquantified, as is our understanding of how age/maturity and sex may mediate their social behaviour. Crucially, the rate of female dispersal between communities is completely unquantified. A much better understanding of bat spatial and social dynamics is necessary to inform statutory functions, effective conservation and epidemiological modelling. We have mapped and quantified the spatial and social dynamics of three communities of Natterer’s bats. Uniquely our roost switching data comes from a community roosting entirely in natural roosts. Radio-tracking, ringing and DNA evidence can be combined at one site, whilst ringing and DNA can be combined at two others. In addition, DNA samples from a further two sites could be included to complete the comparison of 183 Natterer’s bats from 5 sites. Microsatellite data (based on 15 markers) was used to describe relatedness at two functional scales (between roosts within a community and between communities). Relatedness and population structure was also compared to home range analysis and roost use to determine if related individuals forage close to each other or share a roosts more frequently than unrelated individuals. Novel descriptions of demographic and epidemiological rates for this species were determined, which has been incorporated into predictive models of how both the community may respond to changes in the environment, or diseases may spread within the community which will help improve bat Conservation. Description: PhD Thesis

2019-01-01T00:00:00Z Bolam, Friederike Charlotte http://theses.ncl.ac.uk/jspui/handle/10443/4389 2019-07-26T14:51:21Z 2018-01-01T00:00:00Z

Title: Addressing uncertainty and limited data in conservation decision-making Authors: Bolam, Friederike Charlotte Abstract: Biodiversity is declining worldwide at alarming rates, through a range of humaninduced changes. At the same time, there are great uncertainties and biases in our understanding of biodiversity that limit our ability to detect changes. New approaches in estimating and managing uncertainty can inform assessments of the status of biodiversity, and identify what actions might be most beneficial. The thesis examines the applications of these methods in diverse contexts that are of importance to conservation and in which there is limited data available. The potential for Value of Information method to contribute to the prioritisation of conservation action was explored (chapter 2). While its use is increasing, there are currently substantial gaps in its application. Probabilistic graphical models (Bayesian Networks) were built with different Machine Learning algorithms to predict the Red List status of plants, both in the Caatinga region in Brazil (chapter 3) and globally (chapter 4) and to assess why some tiger reserves contain higher tiger numbers than others (chapter 5). Red List status of plants could be predicted reliably by using the number of herbarium specimens of each plant species. The method was used to predict which plants might be threatened globally. The number of poached tigers was a good indicator for the number of tigers in a tiger reserve, but a lack of data at similar spatial scales across the tigers’ range inhibits decision making. Overall, the thesis suggests that we can: a) better predict which species are threatened and prioritise these species for future Red List assessments; b) standardise our research approaches using core outcomes; and c) make better decisions despite uncertainty. We need to make better use of these methods and the currently available data to prevent species from going extinct and to meet global targets aimed to halt the biodiversity crisis. Description: PhD Thesis

2018-01-01T00:00:00Z Peppa, Maria Valasia http://theses.ncl.ac.uk/jspui/handle/10443/4115 2018-12-12T14:21:33Z 2018-01-01T00:00:00Z

Title: Morphology-based landslide monitoring with an unmanned aerial vehicle Authors: Peppa, Maria Valasia Abstract: Landslides represent major natural phenomena with often disastrous consequences. Monitoring landslides with time-series surface observations can help mitigate such hazards. Unmanned aerial vehicles (UAVs) employing compact digital cameras, and in conjunction with Structure-from-Motion (SfM) and modern Multi-View Stereo (MVS) image matching approaches, have become commonplace in the geoscience research community. These methods offer a relatively low-cost and flexible solution for many geomorphological applications. The SfM-MVS pipeline has expedited the generation of digital elevation models at high spatio-temporal resolution. Conventionally ground control points (GCPs) are required for co-registration. This task is often expensive and impracticable considering hazardous terrain. This research has developed a strategy for processing UAV visible wavelength imagery that can provide multi-temporal surface morphological information for landslide monitoring, in an attempt to overcome the reliance on GCPs. This morphological-based strategy applies the attribute of curvature in combination with the scale-invariant feature transform algorithm, to generate pseudo GCPs. Openness is applied to extract relatively stable regions whereby pseudo GCPs are selected. Image cross-correlation functions integrated with openness and slope are employed to track landslide motion with subsequent elevation differences and planimetric surface displacements produced. Accuracy assessment evaluates unresolved biases with the aid of benchmark datasets. This approach was tested in the UK, in two sites, first in Sandford with artificial surface change and then in an active landslide at Hollin Hill. In Sandford, the strategy detected a ±0.120 m 3D surface change from three-epoch SfM-MVS products derived from a consumer-grade UAV. For the Hollin Hill landslide six-epoch datasets spanning an eighteen-month duration period were used, providing a ± 0.221 m minimum change. Annual displacement rates of dm-level were estimated with optimal results over winter periods. Levels of accuracy and spatial resolution comparable to previous studies demonstrated the potential of the morphology-based strategy for a time-efficient and cost-effective monitoring at inaccessible areas. Description: PhD Thesis

2018-01-01T00:00:00Z