Masters theses

Explainable Heterogeneity? A meta-analysis of arctic and subarctic responses to climate change


Species may respond to climate change in a number of ways (Walther et al. 2002) and broad meta-analyses suggests that these responses are occurring globally (Parmesan & Yohe 2003). The way that species respond will have a substantial effect on their survival and the maintenance of regional community assemblages. Distributional responses, for example, are limited by geographic and topographic constraints, and where the ranges of suitable habitat become constricted, species loss may result (Thomas et al. 2004). Thus it is important for conservation planning to be able to predict these responses: however, there is considerable uncertainty in predictions due to substantial heterogeneity in species responses. This study focuses on observed species responses in the arctic and subarctic regions, and uses meta-analysis to explore drivers of heterogeneity in responses. While the results show that species are clearly responding to climatic change in the arctic, there is also considerable heterogeneity in the magnitude of these responses. Methodological and other study-level variation accounted for much of this heterogeneity, while effects of ecological factors were idiosyncratic: spatial location and degree of climate change had no significant effect, and there were only a few trends in the dataset that were explained by habitat or taxonomic factors. Substantial heterogeneity was unexplainable with the collected data, and while there are limitations of the current literature, this may have negative implications for the predictability of future species responses.

View full pdf. Supervised by Dr K. Blake Suttle, Imperial College (Silwood Park), also see my page on the Suttle lab web site.

The effects of small-scale disturbance on invertebrates in a tropical montane cloud forest


Tropical rainforests are under threat from increasing rates of deforestation (Sala et al. 2000, Wright and Muller-Landau 2006), and these habitats contain high levels of biodiversity, endemism and ecosystem functioning (reviews by Gardner et al. 2009, Lewis 2009). The invertebrate fauna is highly diverse and functional within these habitats, and this fauna is both crucial for ensuring the survival of forest processes and also useful in conservation as indicators (Kremen et al. 1993, Spector and Forsyth 1998). While much research has studied the effects of broad-scale disturbance on invertebrates, few studies explore how the communities of invertebrates respond to small-scale disturbance, as this presents challenges in interpretation of results and mediation by environmental variables.

This study focuses on two invertebrate communities, the dung beetles (Scarabaeinae) and orchid bees (Euglossini) in disturbed patches in a tropical montane cloud forest in Honduras. Attractive trapping for these communities was carried out across sites encompassing a range of disturbance properties, and environmental variables for sample sites were also collected. Detrended Correspondence Analysis was carried out to form variables describing community change, and principal coordinates of neighbour matrices used to provide spatially-structure control variables. GLMMs were used to analyse the responses of species richness, total abundance and DCA axes to disturbance, taking into account environmental variation.

Findings showed that both groups display responses to disturbance, and that shifts in community composition appear to drive abundance and richness. The Scarabaeine community shifts from specialist to generalist species in response to increased intensity of disturbance and land use, and Euglossini shift from small-bodied to large-bodied species. Patterns of response in community parameters vary between the two taxa, but both respond relatively rapidly to disturbance. These suggest that ecosystem functioning by these communities will shift in focus, although predictions of rate of change are challenging. Both groups may be good indicators for habitat quality, but their low concordance with one another suggests they may be poor indicators for the wider invertebrate community. Small-scale disturbance research is feasible, and a framework for carrying out this research is presented.

Ongoing research, pdf not available (but contact me if you want more information). Supervised by Dr Robert M. Ewers, Imperial College (Silwood Park), and in association with Operation Wallacea and the Oxford University Museum of Natural History.

Comments are closed.