Viernes, 18 Abril 2014

Modelos predictivos de Tenerife


Elaboración de modelos predictivos de distribución con los datos biológicos almacenados en el Banco de Biodiversidad de Canarias y evaluación de la capacidad de los espacios naturales protegidos de la isla de Tenerife a la hora de representar la diversidad biológica.




Well surveyed squares

 This project aims to analyze the information available of a well-known group (seed plants) in a database (BIOTA-Canarias) which exhaustively compile the presence of species in the Canary Islands.

We specifically work in the Tenerife Island at a pixel resolution of 500 x 500 meters (1,131 species and 1,084,971 database records).



environmental regionalization

Until now we have examined the main shortfalls of this vast amount of information in order to offer reliable estimations of diversity and species richness patterns. Specifically we study the (i) inadequacy of raw data to describe richness patterns due to sampling bias, (ii) lack of survey effort assessment (and lack of exhaustiveness in compiling data about survey effort), and (iii) lack of coverage of the geographic and environmental variations that affect the distribution of organisms.



Limitations of Biodiversity Databases: Case Study on Seed-Plant Diversity in Tenerife, Canary Islands


∗Departamento de Biodiversidad y Biolog´ıa Evolutiva, Museo Nacional de Ciencias Naturales (CSIC), C/Jos´e Guti´errez Abascal, 2,
Madrid 28006, Spain
†Center for Macroecology, Institute of Biology, University of Copenhagen, Universitetsparken 15, DK-2100 Copenhagen O, Denmark
‡Departamento de Ciˆencias Agr´arias, CITA-A, Universidade dos A¸cores, Campus de Angra, Terra-Ch˜a, Angra do Hero´ısmo, 9701-851
Terceira (A¸cores), Portugal


biota tenerife_clip_image002biota tenerife_clip_image002_0000biota tenerife_clip_image002_0001biota tenerife_clip_image002_0002


Species richness

Endemic species


Introduced species

Now we modellize each one of the plant species (more than 800) in order to assess the geographical and biological bias derived from the overlap of many individual models.