Quantitative palaeoclimatic inference based on terrestrial mammal faunas

Abstract

Aim. The main goal of this paper is to propose a new tool, in the form of Quantitative Bioclimatic Models (QnBMs), for reconstructing past climates based on fossil mammal associations. As a case study, European climatic conditions during the basal early Pleistocene will be inferred using QnBMs. Location. The study uses faunas throughout the world to develop a quantitative bioclimatic model, which is then applied to Quaternary faunas from Eurasia. Methods. The models were constructed by applying multivariate linear regression to modern mammal faunas and climates from all over the world. The models were validated with a second group of modern faunas, which includes several from transitional zones between different climates (ecotones). To test the reliability of the method when applied to fossil associations, the results obtained for the Pleistocene have been compared with those obtained from palynology. Results. Validation of the models shows that as many as 11 climatic factors can be inferred with high reliability using the regression models developed in this work. Comparisons of results in the late Pleistocene–Holocene of Barová (Czech Republic) between a palynological study and the quantitative bioclimatic analysis show a high degree of similarity. The results for the early Pleistocene show colder and drier climatic conditions for Europe than today. Main conclusions. The application of the quantitative bioclimatic models to present day as well as Quaternary mammal faunas proves to be a useful tool for palaeoclimatic reconstruction during the Quaternary and probably most of the Neogene. Transfer functions are presented for a complete set of climatic factors, allowing a precise estimation of the climate in a locality from its mammal fauna.