My research aims broadly to understand how human activities are altering the structure and diversity of ecological communities. My current work falls within a number of (often overlapping) research projects.

Quantifying and Predicting Land-use and Climate Effects on Biodiversity

Land use and climate are likely to present the two greatest pressures on biodiversity in the coming decades, but our understanding of their joint and interacting effects is still very limited. I am currently leading a project that seeks to quantify the interacting effects of land use and climate on biodiversity, and to make predictions about the future effects of these pressures.

More details on this project will follow as it develops over the coming months/years.

The Dynamics of African Ecosystems Under Multiple Human Pressures

Africa is a particularly interesting case study for trying to understand human effects on ecological communities for several reasons. First, it is likely to experience the effects of climate change relatively soon (Mora et al., 2013). Second, it is anticpated that much of the agricultural expansion needed to meet increasing human food demands will happen in Africa (Roxburgh et al., 2010). Third, harvesting of wild animals for bushmeat occurs on a large scale in Africa (Milner-Gulland & Bennett, 2003). However, across the whole of Africa, data on the structure of ecological communities is sparse and patchy (Hudson et al., 2014; Meyer et al., 2015). Therefore, although certain African ecosystems are very well understood (Sinclair & Norton-Griffiths, 1979), predicting the effects of human activities across the whole of the continent remains a challenge.

In this project, Ben Collen, Lizzie Boakes and I are making improvements to the Madingley General Ecosystem Model (see below) to make it better able to represent the effects of human pressures on ecosystems. We will then use the improved model to make predictions about the past and future effects of land use, climate and bushmeat hunting on the structure of ecological communities. In order to test whether our predictions agree with observed changes, we will be comparing our results to those inferred from the PREDICTS data (see below) and also to observed population trends from the Living Planet database. We have just published the first paper from this project, comparing predictions of the status and future of biodiversity in African tropical grasslands and savannas made by models from the PREDICTS Project, and the current, unimproved version of the Madingley Model (Newbold et al., in press).

Projecting Responses of Ecological Diversity In Changing Terrestrial Systems (PREDICTS)

Before moving to UCL, I worked full time on the PREDICTS Project, and I am still involved as project partner. In the first phase project, we compiled a global database (now publicly available) of observed land-use effects on species in ecological communities. The database currently contains over 3 million records, from nearly 30,000 locations on the planet, for more than 50,000 species including invertebrates, vertebrates, plants and fungi (Hudson et al., 2014). These data were used to generate global models relating biodiversity to land use (Newbold et al., 2014b; Newbold et al., 2015). By projecting the models onto global maps of land use (historical, current and under future scenarios), we have made predictions about the status and possible futures for the diversity of ecological communities (Newbold et al., 2015; Newbold et al., 2016). We also conducted analyses showing the effects of land use on the turnover in species composition among land uses (Newbold et al., 2016), showing that protected areas are effective in conserving local species diversity even where humans use the land for agriculture (Gray et al., 2016), and showing that traits of bee species influence their responses to land use (De Palma et al., 2015; see also below).

The Madingley Model

A few years ago, I was part of the team that developed the first global General Ecosystem Model (Purves et al., 2013; Harfoot et al., 2014). This is a mechanistic, individual-based model of the dynamics of ecological communities. It represents all plants, and all animal organisms larger than 10 μg. There are several completed (but as yet unpublished) and ongoing analyses using the Madingley Model to understand and predict the effects of human activities on the structure and dynamics of ecosystems (see also Dynamics of African Ecosystems Project, above).

Species Traits and Responses to Human Pressures

A lot of my work over the past few years has investigated how the ecological characteristics of species influence the way that they respond to environmental change driven by human pressures. One piece of work showed that for bird species in tropical forests, large-sized, slow-breeding, non-migratory forest specialists with diets of fruits and invertebrates respond to more to land use than other species (Newbold et al., 2013). By projecting these models onto maps of land use, we estimated the consequences of trait-mediated responses to land use for the diversity and functional structure of forest bird communities across the tropics (Newbold et al., 2014a). I have also done some work trying to answer similar questions - but for more species than just birds - using the PREDICTS data. So far the only published work is that showing that traits of bees influence their responses to land use (De Palma et al., 2015), but there are more analyses ongoing.

▼ Past Projects ▼

Species Distribution Models as a Tool for Conservation

In my PhD I investigated the use of species distribution models for guiding conservation, particularly in the context of Egypt. At the time one of my PhD supervisors, Francis Gilbert, was on sabbatical in Egypt running - together with Samy Zalat - the BioMAP Project, which was collating recorded sightings of species in Egypt. I used these data to develop distribution models, which I used to estimate patterns of species richness (Newbold et al., 2009a), and to show that ecological characteristics of butterfly species influence how well their distributions correlate with climatic conditions (Newbold et al., 2009b). I also conducted fieldwork to show that the distribution models gave a reasonable estimation of species' distributions (Newbold et al., 2010), and wrote a review of the use of museum data for modelling species' distributions (Newbold, 2010).