A carregar...
Publicação
Mammals and where to find them: biodiversity patterns and data curation challenges
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 13.77 MB | Adobe PDF |
Orientador(es)
Resumo(s)
Biodiversity is life’s foundation, key to ecosystem services and functioning, yet threatened, and requiring protection. Ecological data, though often scarce or imperfect, offer precious insights. In this work, occurrence data for twelve mammal species, from over a decade of track transects and camera trapping, were used to model biodiversity drivers and species distribution, and identify areas suitable for conservation within the study area, Portugal’s largest agro-silvo-pastoral estate. This extensive, flawed dataset required considerable data curation. The data were compiled, selected, pre-processed and error-checked using custom R scripts, and uploaded to a PostgreSQL database designed and created for this purpose. Multiple model sets were then created, using different methods (GLMs, GAMs, MaxEnt) and data, to identify the effects of several variables on each species. From these, coefficient tables were created, allowing in-depth variable effect analyses, plus multiple sets of predicted distribution maps for all species, facilitating future project planning. These sets were merged into biodiversity (Shannon- Wiener) index maps, then averaged into a single biodiversity index map, and, considering economic suitability, used to create a conservation suitability map. A broad area suitable for conservation was identified, to later inform the creation of smaller conservation areas – expected to provide valuable biodiversity protection – alongside stakeholders, considering knowledge and data not available here. Concurrently, based on patterns identified during data compilation, data management guidelines were developed, to help improve and standardize future data collection and handling, allowing greater robustness. This work pioneered data management efforts in the study area, and identified the effects of an extensive set of variables on multiple species, at a scale unprecedented for the study area, allowing better-informed management and conservation actions. Likewise, the resulting dedicated database will hopefully accommodate the compilation of the remaining data archives, alongside data from ongoing research, multiplying the impact of future research and management efforts.
Descrição
Tese de Mestrado, Biologia da Conservação, 2025, Universidade de Lisboa, Faculdade de Ciências
Palavras-chave
Biodiversity assessment, biodiversity drivers protected areas databases data processing