Aim：To understand the scenarios of ‘anthropogenic biomes’ that integrate human and ecological systems, we need to explore the impacts of climate and human disturbance on vegetation in the past and present. Interactions among surface pollen, modern vegetation and human activities along climate and land-use gradients are tested to evaluate the natural and anthropogenic forces shaping the modern vegetation, and hence to aid the reconstruction of vegetation and climate in the past. This in turn will help with future predictions.

Location：The North-east China Transect (NECT) in north-eastern China.

Methods：We analysed 33 surface pollen samples and 213 quadrats across four vegetation zones along the moisture/land-use gradients of the NECT. Detrended correspondence analysis (DCA) and redundancy analysis (RDA) of 52 pollen taxa and three environmental variables were used to distinguish anthropogenic and climatic factors that affect surface pollen assemblages along the NECT.

Results：The 33 surface samples are divided into four pollen zones (forest, meadow steppe, typical steppe and desert steppe) corresponding to major vegetation types in the NECT. Variations in pollen ratios of fern/herb (F/H), Artemisia/Chenopodiaceae (A/C) and arboreal pollen/non-arboreal pollen (AP/NAP) represent the vegetation and precipitation gradient along the NECT. DCA and RDA analyses suggest that surface pollen assemblages are significantly influenced by the precipitation gradient. Changes in the abundance of Chenopodiaceae pollen are related to both human activities and precipitation.

Main conclusions：Surface pollen assemblages, fossil pollen records, archaeological evidence and historical documents in northern China show that a large increase of Chenopodiaceae pollen indicates human-caused vegetation degradation in sandy habitats. The A/C ratio is a good indicator of climatic aridity, but should be used in conjunction with multiple proxies of human activities and climate change in the pollen-based reconstruction of anthropogenic biomes.

Key words: Climate gradient; human activities; North-east China Transect (NECT); ordination; redundancy analysis (RDA); surface pollen