Relative pollen productivity estimates for major plant taxa of cultural landscapes in central eastern China

Qinghai Xu,Dominique Laffly,Florence Mazier,Shinya Sugita,Yuyun Zhang,Furong Li,Zhongze Zhou,Marie-José Gaillard,Yuecong Li

doi:10.1007/s00334-017-0636-9

Abstract

In this study we estimate relative pollen productivity (RPP) for plant taxa characteristic of human-induced vegetation in ancient cultural landscapes of the low mountain ranges of Shandong province in eastern temperate China. RPP estimates are required to achieve pollen-based reconstructions of Holocene plant cover using modelling approaches based on Prentice’s and Sugita’s theoretical background and models (REVEALS and LOVE). Pollen counts in moss samples and vegetation data from 36 sites were used in the Extended R-Value (ERV) model to estimate the relevant source area of pollen (RSAP) of moss polsters and RPP of major plant taxa. The best results were obtained with the ERV sub-model 3 and Prentice’s taxon-specific method (using a Gaussian Plume dispersal model) to distance weight vegetation data. RSAP was estimated to 145 m using the maximum likelihood method. RPP was obtained for 18 taxa of which two taxa had unreliable RPP (Amaranthaceae/Chenopodiaceae and Vitex negundo). RPPs for Castanea, Cupressaceae, Robinia/Sophora, Aster/Anthemis-type, Cannabis/Humulus, Caryophyllaceae, Brassicaceae and Galium-type are the first ones for China. Trees, except Robinia/Sophora (RPP = 0.78 ± 0.03) have larger RPPs than herbs other than Artemisia (RPP = 24.7 ± 0.36). The RPPs for Quercus, Pinus and Artemisia are comparable with other RPPs obtained in China, the RPPs for Pinus, Quercus, Ulmus, Cyperaceae and Galium-type with the mean RPPs obtained in Europe, and RPP for Cupressaceae with that for Juniperus in Europe. The values for Aster/Anthemis-type, Caryophyllaceae, Asteraceae SF Cichorioideae and Juglans differ from the few RPPs available in China and/or Europe.

Highlights

Quantitative reconstructions of anthropogenic land-cover change at the regional to continental spatial scale, especially the relative size and distribution of open and wooded land area, are of great value in answering research questions related to climate change (e.g. Gaillard et al 2010; Strandberg et al 2014)
The impact of vegetation survey methods on the relevant source area of pollen (RSAP) sensu Sugita (1994) and relative pollen productivity (RPP) estimates was discussed in Bunting and Hjelle (2010) and a standardized vegetation-survey protocol was proposed by Bunting et al (2013b)
The relatively high number of RPP estimates available in Europe made it possible to test and use the models developed for quantitative reconstruction of vegetation cover such as the Landscape Reconstruction Algorithm (LRA: REVEAL and LOVE models) (Sugita 2007a, b) and the Multiple Scenario Approach (MSA; Bunting and Middleton 2005)

Summary

Introduction

Quantitative reconstructions of anthropogenic land-cover change at the regional to continental spatial scale, especially the relative size and distribution of open and wooded land area, are of great value in answering research questions related to climate change (e.g. Gaillard et al 2010; Strandberg et al 2014). We estimate relative pollen productivity (RPP) of major tree and herb taxa in ancient agrarian landscapes of central and southern Shandong, central eastern China (Fig. 1), as the first step of a larger project using RPPs and Sugita’s REVEALS model (Sugita 2007a) to achieve pollen-based quantitative reconstructions of Holocene regional vegetation cover. The relatively high number of RPP estimates available in Europe made it possible to test and use the models developed for quantitative reconstruction of vegetation cover such as the Landscape Reconstruction Algorithm (LRA: REVEAL and LOVE models) (Sugita 2007a, b) and the Multiple Scenario Approach (MSA; Bunting and Middleton 2005). The MSA was applied in northern Sweden (Karlsson et al 2008)

Methods

Results

Discussion

Conclusion