作者: Fang Tian , Xianyong Cao , Anne Dallmeyer , Yan Zhao , Jian Ni
DOI: 10.1016/J.QUASCIREV.2016.11.027
关键词:
摘要: Abstract Temporal and spatial stability of the vegetation–climate relationship is a basic ecological assumption for pollen-based quantitative inferences past climate change predicting future vegetation. We explore this Holocene in eastern continental Asia (China, Mongolia). Boosted regression trees (BRT) between fossil pollen taxa percentages ( Abies , Artemisia Betula Chenopodiaceae, Cyperaceae, Ephedra Picea Pinus Poaceae Quercus ) model outputs mean annual precipitation (P ann temperature warmest month (Mt wa 9 6 ka (ka = thousand years before present) were set up results compared to those obtained from relating modern climate. Overall, our reveal only slight temporal differences pollen–climate relationships. Our analyses suggest that importance P with Mt distribution higher today than it was at ka. In particular, relevance increases has become main determinant. This climate–tree parallels widespread tree decrease north-central China Tibetan Plateau. assume least partly related disequilibrium originating human impact. Increased atmospheric CO 2 concentration may have permitted expansion moisture-loving herb (Cyperaceae Poaceae) during late into arid/semi-arid areas. furthermore find Plateau generally similar, but regional are larger differences. summary, relationships stable space time, reconstructions can be applied Holocene. Regional imply calibration-set should restricted spatially.