Combining distribution modelling and non-invasive genetics to improve range shift forecasting

• Type: Article in International Scientific Journal
• Date: 2015

Authors

Frederico Mestre (Author) Other; Ricardo Pita (Author) Other; Joana Pauperio (Author) Other; Filipa M S Martins (Author) Other; Paulo Cello Alves (Author) FCUP; Antonio Mira (Author) Other; Pedro Beja (Author) Other

Journal

Title: Ecological Modelling

Vol. 297

Pages: 171-179

ISSN: 0304-3800

Publisher: Elsevier

Indexing

ISI Web of Knowledge - 15 Citations

Scopus - 17 Citations

Scientific classification

FOS: Natural sciences > Biological sciences

Other information

Authenticus ID: P-009-ZXZ

DOI: 10.1016/j.ecolmodel.2014.11.018

Abstract (EN): Forecasting species range shifts under climate change is critical to adapt conservation strategies to future environmental conditions. Ecological niche models (ENMs) are often used to achieve this goal, but their accuracy is limited when species niches are inadequately sampled. This problem may be tackled by combining ENM with field validation to fine-tune current species distribution, though the traditional methods are often time-consuming and the species ID inaccurate. Here we combine ENM with novel field validation methods based on non-invasive genetic sampling to forecast range shifts in the globally near-threatened Cabrera vole (Microtus cabrerae). Using occurrence records mapped at 10 km x 10 km resolution, we built the first ENM (ENM1) to estimate the current species distribution. We then selected 40 grid squares with no previous data along the predicted range margins, and surveyed suitable habitats through presence-sign searches. Faecal samples visually assigned to the species were collected for genetic identification based on the mitochondrial cytochrome-b gene, which resulted in 19 new grid squares with confirmed presence records. The second model (ENM2) was built by adding the new data, and species distribution maps predicted by each model under current and future climate change scenarios were compared. Both models had high predictive ability, with strong influence of temperature and precipitation. Although current distribution ranges predicted by each model were quite similar, the range shifts predicted under climate change differed greatly when using additional field data. In particular, ENM1 overlooked areas identified as important by ENM2 for species conservation in the future. Overall, results suggest that combining ENM with non-invasive genetics may provide a cost-effective approach in studies regarding species conservation under environmental change.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 9