Modern vegetation–pollen–climate relationships for the Pampa grasslands of Argentina

Aim To analyse the relationships between potential natural vegetation, pollen and climate in order to improve the interpretation of fossil pollen records and provide the background for future quantitative palaeoclimatic reconstructions. Location Pampa grasslands of Argentina, between 33–41° S and 56–67° W. Methods Modern pollen data were obtained from a pollen data base developed by the Grupo de Investigación de Paleoecología y Palinología, Universidad Nacional de Mar del Plata, Argentina (143 surface samples and 17 pollen types). Analysis of pollen and climate data involved multivariate statistics (cluster analysis and principal components analysis), scatter diagrams, Pearson’s correlation and isopoll mapping. Results Vegetation patterns at regional scales (grasslands and xerophytic woodlands) and local scales (edaphic communities) were identified by cluster analysis of pollen surface samples. The main climatic variables that appear to constrain the vegetation distribution and abundance of taxa are mean annual precipitation, annual effective precipitation and summer temperature. Individual pollen types such as Chenopodiaceae, Apiaceae, Cyperaceae, Prosopis, Schinus, Condalia microphylla and other xerophytic taxa are good indicators of moisture regime. Many pollen types are significantly correlated with summer temperature. The modern vegetation–pollen–climate relationships vary in a broadly predictable manner, supporting the contention that fossil pollen assemblages can be related to particular climatic characteristics. Main conclusions An expanded suite of modern analogues facilitated new insights into vegetation–pollen–climate relationships at the regional scale in Pampa grasslands. Relationships between individual pollen types and climate are appraised at a regional scale and new modern analogues are presented. The results provide the basis for improved vegetation and climate reconstruction from fossil records of the study area.