Use this URL to cite or link to this record in EThOS:
Title: Extrapolating insect biodiversity across spatial scales
Author: Barwell, Louise Joanne
ISNI:       0000 0004 5918 1897
Awarding Body: University of Leeds
Current Institution: University of Leeds
Date of Award: 2015
Availability of Full Text:
Access from EThOS:
Access from Institution:
Extrapolating biodiversity patterns across spatial scales can address shortfalls in our knowledge of species distributions, inform conservation and further our understanding of spatial patterns in biodiversity. I compared fine grain predictions of occupancy for British Odonata species among ten downscaling models. I observed a sigmoidal occupancy-area relationship for the best performing model and found that predictive success for Odonata species varied systematically with species traits. Species with high dispersal abilities had greater predictive error. Poorer predictions for species with a climatic range limit in Britain suggest that environmental information is required to fully capture spatial patterns in biodiversity. I modelled the distribution of the Brindled Green moth at two spatial grains using a hierarchical Bayesian model to quantify associations with climate, landcover and elevation, whilst accounting for residual spatial autocorrelation and spatial patterns in recording effort. Model predictions improved at the finer spatial grain and identified unsurveyed grid cells with high suitability for future recording. The overlap between individual species distributions underpins spatial patterns in multi-species assemblages. I used simulated species assemblages to evaluate 29 abundance-based metrics of β-diversity against a set of desirable and ‘personality’ properties. Metrics accounting for unseen shared and unshared species were lacking. I identified a trade-off between robustness in the face of undersampling and sensitivity to turnover in rare species. The findings were borne out when a selection of metrics were applied to assemblages of British macro-moths: variation in β-diversity was best explained by climate, landcover and distance when using standardised data and abundance-based metrics, as opposed to opportunistic data and presence-absence metrics. This thesis has demonstrated the value of using biological records to explore biodiversity patterns at multiple spatial scales and has highlighted some of the methodological challenges that remain.
Supervisor: Kunin, William E. ; Isaac, Nick J. B. Sponsor: NERC
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available