Trends and fluctuations in species abundance and richness in Mediterranean annual pastures
Title | Trends and fluctuations in species abundance and richness in Mediterranean annual pastures |
Publication Type | Journal Article |
Year of Publication | 1998 |
Authors | Peco, B., Espigares T., & Levassor C. |
Journal | Applied Vegetation Science |
Volume | 1 |
Pagination | 21-28 |
Keywords | Autumn rainfall distribution, Generalized Linear, Models, Post-ploughing succession, Topography |
Abstract | Floristic data collected from permanent plots during 14 consecutive years are used to model the frequency of the 62 most abundant species in relation to post-ploughing succession, topography and rainfall in annual Mediterranean grasslands located in a Quercus rotundifolia dehesa. The interannual dynamics of species richness are also analysed. From 1980 to 1993, presence/absence data of grassland species were noted in five 20 cm × 20 cm permanent quadrats placed at random in 1980 in 14 permanent plots on a southfacing slope along the topographic gradient. Weekly autumn rainfall data over the 14 years were analysed using a profile attributes index and Hybrid Multidimensional Scaling to arrange the years according to their autumn rainfall pattern. Generalized Linear Models were used to fit the species richness and species frequency according to topographic position, age since the last ploughing episode, total rainfall in the growing season and autumn rainfall pattern using a forward stepwise procedure. The richness model includes all of these variables, and reveals a relatively high goodness-of-fit (71%). The fact that the meteorological factors play a key role in modelling richness forces us to include them if we wish to use richness as an indicator of the degree of disturbance in these highly fluctuating annual pastures. Models of species dynamics show that although roughly 33% of the species have a successional behaviour, the majority are more dependent on temporal heterogeneity associated with rainfall or spatial heterogeneity linked to the topographic gradient. |