The effect of urbanisation on biodiversity in space and time

Overview

Urbanisation is one of the most pervasive forms of habitat change. More than half of the world’s human population now resides in urban areas, and urban land cover is projected to triple between 2000 and 2030 (1). Anthropogenically-driven land conversion from a natural or semi-natural state to intensive agriculture and urban built environment poses a major threat not only to particular species, but also to biodiversity (1).
Anthropogenic land use varies considerably in time and space, making it a key challenge to account for both dimensions in our attempts to understand how biodiversity changes as a result of urbanisation (2). Urban sprawl typically results in habitat fragmentation at the transition zone between urban and rural areas. As this process unfolds, two competing forces are at play: habitat fragmentation makes populations less viable, potentially leading to extinctions that negatively impact biodiversity; conversely, the landscape becomes more diverse, potentially opening up new niches at the fringes of urban development. Later, habitat homogenisation through in-filling of previously spared tracts of land would be expected to decrease biodiversity. All these processes can vary in their speed and spatial extent, and can even be reversed if urban areas become depopulated (2). As a result, the extinction of existing species and the colonisation of new niches by opportunist species might generate an increasing, decreasing or hump-shaped relationship between biodiversity and urbanisation (Fig. 2c).
Existing research on the relationship between biodiversity and urbanisation is inconclusive. A substantial body of work has suggested that urbanisation leads to biodiversity loss through biotic homogenisation of animal communities (3, 4). Some of these studies provide evidence for a hump-shaped relationship with urbanisation (3). More recent work suggests that when green spaces are provided, cities, suburban areas and intensively cultivated rural land can host a considerable proportion of regional biodiversity (5). However, to date, research has focused on rural-urban gradients across space, with little consideration of changes over time (2). This is a critical knowledge gap, potentially leading to biases in the interpretation of previous results, especially if there is a time lag in the response of biodiversity to land use change. Moreover, previous studies have generally been conducted at small spatial scales, making it difficult to generalise observed relationships (but see (5)).
This project will investigate the response of biodiversity – measured for each site individually (alpha diversity), between sites (beta diversity), and regionally (gamma diversity) (6) – as urbanisation has progressed. The integration of data collected over time at a broad biogeographical scale will help distinguish, for the first time, key factors that modulate the rate and directionality of biodiversity change with urbanisation.
A range of factors might be expected to modulate the relationship between urbanisation and biodiversity. For example, different regions may differ in environmental conditions such as climate, altitude, temperature and precipitation, major abiotic drivers of community composition and diversity (7, 8). Areas also differ in their history of land use (e.g. agricultural vs forest land) and such legacies may affect future states of biodiversity (2). Finally, communities vary naturally as a result of dispersal and stochastic demographic processes (9). These lead to variation in community structure and biodiversity, with resultant effects on resistance and/or resilience to land use change that can affect the relationship between urbanisation and biodiversity (8, 10). At larger spatial scales, biotic homogenisation can lead to limited differences between sites (low beta diversity), ultimately impacting regional diversity (gamma diversity) (9, 10).
This project seeks to understand the relationship between biodiversity and urbanisation in space and time. The ultimate aim of this project is to develop quantitative methods that will transform the way we investigate the effect of urbanisation on biodiversity and provide the vital scientific information required to build cities that promote sustainable urban ecosystems. Specifically, the project asks:

1) What are the characteristics of an urban habitat that can sustain high levels of biodiversity?

2) Is there a stage of urban sprawl at which biodiversity declines sharply?

3) How does the past ecological state (climatic, environmental, land use, species richness/evenness) of an area affect its capacity to buffer against urbanisation?

Answering these questions will fundamentally advance our understanding of how anthropogenic habitat change shapes biodiversity. Moreover, we will gain insights into how specific drivers of loss and gain can affect different aspects of biodiversity, which will enable us to make recommendations for urban planning to encourage more sustainable and biodiverse cities (11).

Click on an image to expand

Image Captions

fig 1.png – Fig. 1. Anthropogenic habitat change occurs at different rates and intensities, and takes a range of forms. This results in a mosaic of land use types that changes through time, exerting pressure on local and regional biodiversity. Source: Roschetzky Photography @ BigStockPhoto.com

figure 2.png – Fig. 2. Land-use maps can be obtained at different points over at least a 15-year period. In the example, impervious surface data from the city of Houston, US, is shown for the year 2000 and 2016 (a, b). Change in land cover over time can then be combined with biodiversity data. In the example we show all the survey transects from the US Breeding Bird Survey (c). These data can be used to calculate biodiversity indexes for each different location at each time point. Finally, spatially-explicit models can be built to assess how anthropogenic habitat change relates to changes in biodiversity (d).

Methodology

We will integrate bird biodiversity data collected over at least two decades in the US with environmental and land use data over the same time period. Both datasets are freely available for download, and demonstrates a number of characteristics of relevance for this project: the data cover an extremely large area (given the scale of the US) with large latitudinal and longitudinal gradients, and a substantial period of time during which significant urbanisation has taken place; further, the avian data are based on dense bird monitoring network that passes close to or through urban areas.
We will model both temporal and spatial effects of urbanisation on biodiversity, largely using Bayesian models. This will allow us to fully characterise the roles of time, space and other explanatory variables, permitting future biodiversity projections and interpolations to non-censused areas. Our project will thus help distinguish key properties of the urban environment (e.g. size and composition of urban green areas) that correlate with high biodiversity (objective 1), and whether/when during the urbanisation process biodiversity declines sharply (objective 2). Moreover, areas can differ in their history of land use and ‘original’ biodiversity, which may or may not buffer against urbanisation. For example, more even communities might be expected to be less sensitive to the effects of urbanisation (objective 3). Finally, knowledge from objectives 1 and 2 can be integrated into predictive models that will help identify areas that are at high risk of losing bird diversity in the face of future urban sprawl.

Project Timeline

Year 1

Retrieval and data organisation of avian biodiversity data and land-use and environmental data from online repositories and published work; writing of review paper on the topic; initial analyses for objective 1.

Year 2

Building of spatio-temporal models of changes in bird abundance in relation to changes in land use and environmental factors (objectives 2 and 3); data analysis and manuscript preparation for objective 1; attendance of course on Bayesian analysis of spatio-temporal distribution data; attendance of national scientific meeting (i.e. British Ecological Society), visit to CEH.

Year 3

Finalise analyses and write up for objective 2 and 3; attendance of scientific writing courses, attendance of national scientific meeting.

Year 3.5

Attendance of international scientific meeting, visit to CEH, completion of manuscripts and submission of thesis.

Training
& Skills

The training provided by this project will cover a broad suite of important formal quantitative and computational approaches and their application, providing the student with an extremely strong basis for pursuing independent research in their field(s) of interest or for a transition to roles in growth areas such as data science. The successful candidate will develop interdisciplinary skills at the intersection of mathematics, statistics and ecology. These will include automated processing of citizen science data, as well as statistical and computational methods applied in ecology and epidemiology.
DD is Lecturer in Urban Ecology at IBAHCM. His research focuses on the adaptation of species to city life. He will provide the student with knowledge about the anthropogenic drivers of changes in organismal health as well as community dynamics along urban gradients.
RM is a Leckie Research Fellow, working on urbanisation and ecosystem health, based across IBAHCM and the Social and Public Health Sciences Unit (SPHSU). She has a strong interdisciplinary background, with expertise in mathematical and computational modelling of ecological systems. From RM, the student will acquire knowledge of computational data processing and stochastic dynamic metapopulation modelling, as well as of a suite of newly-developed measures of biodiversity.
SS is Lecturer in Community Ecology at the School of Life Sciences and Associate of IBAHCM. She integrates computational modelling with laboratory and field experiments to unravel mechanisms that promote species coexistence and biodiversity. Through interaction with SS, the student will acquire knowledge on the quantitative assessment and interpretation of different measures of biodiversity.
JM is Professor of Spatial and Population Ecology at IBAHCM. His work focuses on building theory by translating biological hypotheses to mathematical models, fitting these to population and demographic data, and applying the conclusions to wildlife conservation and risk assessment. He will train the student in fitting spatially-explicit metacommunity models within a Bayesian framework.
The scholar will be based within IBAHCM and the award-winning Boyd Orr Centre for Ecosystem and Population Health. As well as selecting from a variety of postgraduate courses for PhD students based on needs, in year 1, the student will receive training on spatial modelling, GIS, diversity indices and management of large datasets by following dedicated courses offered within IBAHCM and through the support of the supervisors. The student will also receive training on using the University of Glasgow compute cluster to enable rapid data processing. In year 2, the student will develop skills in spatio-temporal modelling, and apply these to integrating different datasets on biodiversity and environmental variables. In year 3, the student will join retreat sessions on scientific writing, organised by the IBAHCM PhD cluster, to help with thesis writing and manuscript preparation. Through participation in Institute seminars and national and international conferences, she/he will also develop presentation and communication skills.

References & further reading

1. K. C. Seto et al PNAS 109, 16083-8 (2012).
2. C. E. Ramalho et al Trends Ecol. Evol. 27, 179-88 (2012).
3. M. L. McKinney Biol. Conserv. 127, 247-60 (2006).
4. E. Knop Glob. Chang. Biol. 22, 228-236 (2015).
5. M. F. J. Aronson et al Proc. R. Soc. London B 281, 20133330 (2014).
6. J. B. Socolar et al Trends Ecol. Evol. 31, 67-80 (2016).
7. C. Lamanna et al PNAS 111, 13745-50 (2014).
8. M. Vandewalle et al Biodivers. Conserv. 19, 2921-47 (2010).
9. D. Sol et al Ecol. Lett. 17, 942-50 (2014).
10. H. von Wehrden et al Landsc. Ecol. 29, 941-8 (2014).
11. M. F. J. Aronson et al Front. Ecol. Environ. 15, 189-96 (2017).

Further Information

Applications: to apply for this PhD please use the url: https://www.gla.ac.uk/study/applyonline/?CAREER=PGR&PLAN_CODES=CF18-7316.

This project is in competition with others for funding, and success will depend on the quality of applicants. Funding includes tuition fee waiver for Glasgow University, a competitive stipend, and research support. To express interest please contact Dr Davide Dominoni (Davide.Dominoni@glasgow.ac.uk) or Dr Rebecca Mancy (Rebecca.Mancy@glasgow.ac.uk) by early January 2020, including: 1) a paragraph detailing your reasons for applying and how your experiences fit the project; 2) your CV with marks earned for previous degrees; and 3) contact info for two references. Only the best applicants will be asked to submit a full application, including two reference letters, by 16:00 on the 10th of January 2020.

Apply Now