Post‐glacial patterns in vegetation dynamics in Romania: homogenization or differentiation?

Abstract

AbstractAim This paper examines eight fossil pollen datasets from Romania with the aim of exploring regional and elevational patterns in site similarity throughout the Holocene. In particular, we aim to determine whether there are clear intervals of homogenization/differentiation and to ascertain the potential driving factors.Location Romania.Methods Qualitative (pollen diagrams) and numerical methods including principal components analysis and Bray–Curtis similarity analyses were used.Results We found strong variability in the past vegetation dynamics during the Holocene. Bray–Curtis similarity analyses show large fluctuations in vegetation similarity and distinct periods of homogenization and differentiation throughout the Holocene. The magnitude and length of these periods appear quite variable in time, but the significant ones can be delimited as follows: (1) differentiation between 11,250 and 11,000 cal. yr bp, 10,000 and 9750 cal. yr bp, 6000 and 5750 cal. yr bp, 2500 and 2250 cal. yr bp, and especially over the last 200 years; and (2) homogenization between 9750 and 9500 cal. yr bp, and 2750 and 2500 cal. yr bp, with more stable periods between 9000 and 7750 cal. yr bp, 4750 and 3500 cal. yr bp, and 2000 and 1000 cal. yr bp.Main conclusions First, periods of biotic homogenization that occurred before significant anthropogenic impact on vegetation demonstrate that not all homogenization is a product of anthropogenic change: it can also be driven by natural causes. In fact, recent human impact (over the last 200 years) appears to have resulted in increased regional differentiation and not in homogenization – a result that contradicts most studies based on more modern, short‐term records. Second, both abiotic (climate and disturbance) and biotic factors are likely drivers of intervals of differentiation and homogenization. We suggest that differentiation may be triggered primarily by climate changes and disturbances (mostly natural pre‐2500 cal. yr bp and human‐induced thereafter), whereas homogenization may be driven predominantly by biotic interactions (e.g. immigration and interspecific competition). Third, this long‐term study raises awareness that assessments of pattern in vegetation homogenization/differentiation may depend on the specific time period and length of investigation. Long‐term investigations through multiple generations are likely to yield particularly useful information on the mechanisms and effects of biotic homogenization.