The LGM Distribution of Dominant Tree Genera in Northern China's Forest-steppe Ecotone and Their Postglacial Migration

Springer Theses Recognizing Outstanding Ph.D. Research

Qian Hao

The LGM Distribution of Dominant Tree Genera in Northern China’s Forest-steppe Ecotone and Their Postglacial Migration

Springer Theses Recognizing Outstanding Ph.D. Research

Aims and Scope The series “Springer Theses” brings together a selection of the very best Ph.D. theses from around the world and across the physical sciences. Nominated and endorsed by two recognized specialists, each published volume has been selected for its scientific excellence and the high impact of its contents for the pertinent field of research. For greater accessibility to non-specialists, the published versions include an extended introduction, as well as a foreword by the student’s supervisor explaining the special relevance of the work for the field. As a whole, the series will provide a valuable resource both for newcomers to the research fields described, and for other scientists seeking detailed background information on special questions. Finally, it provides an accredited documentation of the valuable contributions made by today’s younger generation of scientists.

Theses are accepted into the series by invited nomination only and must fulfill all of the following criteria • They must be written in good English. • The topic should fall within the confines of Chemistry, Physics, Earth Sciences, Engineering and related interdisciplinary fields such as Materials, Nanoscience, Chemical Engineering, Complex Systems and Biophysics. • The work reported in the thesis must represent a significant scientific advance. • If the thesis includes previously published material, permission to reproduce this must be gained from the respective copyright holder. • They must have been examined and passed during the 12 months prior to nomination. • Each thesis should include a foreword by the supervisor outlining the significance of its content. • The theses should have a clearly defined structure including an introduction accessible to scientists not expert in that particular field. More information about this series at http://www.springer.com/series/8790

Qian Hao

The LGM Distribution of Dominant Tree Genera in Northern China’s Forest-steppe Ecotone and Their Postglacial Migration Doctoral Thesis accepted by Peking University, Beijing, China

Qian Hao Institute of Surface-Earth System Science Tianjin University Tianjin, China

Supervisor Hongyan Liu College of Urban and Environmental Sciences Peking University Beijing, China

ISSN 2190-5053     ISSN 2190-5061 (electronic) Springer Theses ISBN 978-981-13-2882-4    ISBN 978-981-13-2883-1 (eBook) https://doi.org/10.1007/978-981-13-2883-1 Library of Congress Control Number: 2018958510 © Springer Nature Singapore Pte Ltd. 2018 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Supervisor Foreword

Forests account for 30% of the global land area, and they have important regulatory roles in the water cycle and biogeochemical cycle of the entire earth. However, climate change has led to increasing forest mortality rates worldwide; thus, the forest dynamic is an important ecology issue under future climate change, which needs long-term study. In her thesis study, Dr. Qian Hao selected the forest-steppe ecotone in China as the main study areas, where the vegetation is sensitive to climate change. She has combined different approaches of paleoecology, phylogeography, and ecological modeling to reconstruct the glacial refugia and postglacial migration of three dominate forest genera (Pinus, Quercus, and Betula) in northern China. Through the studies of different temporal and spatial scales, she got three main conclusions: (1) the postglacial migration was multidirectional, including not only the long-distance migration process from low to high latitudes but also the process of local diffusion from northern refugia; (2) except the latitudinal migration, the vertical migration altered the changes of forest-steppe border to the Holocene climate changes; (3) there were various responses of local forests to climate change, including decreasing forest coverage, shrub expansion, and replacement of steppe. In sum, the results and major findings of the study can help us to understand the forest migration patterns comprehensively and the factors influencing the forest dynamics. This study will be useful for the fields of forest ecology, paleoecology, and biogeography. College of Urban and Environmental Sciences  Peking University, Beijing, China August 2018

Hongyan Liu

v

Preface

Studying dynamics of forests that occupy 30% of the land surface area has been becoming a hotspot in ecology. Forest distribution change since the Last Glacial Maximum (LGM) is critical for validating models predicting future forest dynamics. On the regional and catchment scale, the different resilience of tree species and buffer effects from diversified land surface features lead to the complex forest distributions and migration patterns in the forest-steppe ecotone, which need to be investigated in a multidisciplinary approach. The forest-steppe ecotone in northern China is located at the edge of East Asian monsoon influence, and the regional forest within the ecotone is sensitive to climate change. Based on high-resolution lacustrine sedimentary records, this study reconstructed the distribution changes of Pinus, Quercus, and Betula during and after the LGM. An interdisciplinary approach integrating methods of paleoecology, phylogeography, and ecological modeling was used to reconstruct their glacial refugia and postglacial migration in northern China. On the regional scale, eight high-­ resolution sediment records under similar climate conditions were selected to show altitudinal migration pattern of these three dominated tree genera and the role of topographical factors on the migration of the forest-steppe border. On the catchment scale, an 820 cm core in Huangqihai, located in the forest-steppe ecotone in northern China, was analyzed in order to specify the local forest dynamics reconstructed from pollen assemblage as response to the Holocene climatic change reconstructed from chemical element ratio in the sediment. The following conclusions are drawn from this study: 1) The postglacial dynamics of forest distribution included not only long-distance north-south migration but also local spread process from LGM refugia in northern China. Our data supported the hypothesis of the mountainous refugia in northern China. Besides, the Otindag Sandy Land which serves as the northern limit of Pinus tabulaeformis distribution was for the first time proposed working as a local refugium. Hence, we suggest that the species was located further north than previously illustrated, albeit perhaps at very low density. We thus infer that postglacial migration of vii

viii

Preface

P. tabulaeformis was multidirectional and implied local spread from various refugia as well as long-distance migration from central/southern China. This study also used the fossil evidence and species distribution model to illustrate the distribution area of other three dominated tree species (Quercus mongolica, Betula platyphylla, and Betula dahurica) in northern China during the LGM. The results also showed that the mountains in northern China may be refugia, and the northern coastal region may also work as refugia, suggesting the existence of the local spread phenomenon for them. 2) The dynamics of forest included not only latitudinal migration but also altitudinal migration. Altitudinal migration altered the changes of forest-steppe border to the Holocene climate changes. Tree species living in the forest-steppe ecotone were commonly controlled by climate, especially the monsoon evolution. However, forests not only migrated latitudinally between northern and southern regions but also showed changes in altitudinal distribution during the Holocene, which is mainly influenced by topographic factors. Areas with high altitude range can support more trees and altitudinal migration possibility. Altitudinal migration altered the changes of forest-steppe border to the Holocene climate changes, and different topographic conditions lead to the spatial heterogeneity. 3) The response modes of local forest to climate drying could be shrubland expansion and savannification other than the replacement of forest by steppe. The Holocene climatic change of Huangqihai region captured in detail the regional-scale monsoon dynamics, especially the drying trend which was obvious during 4000 to 1000 cal year BP.  The corresponding vegetation dynamics were characterized by a steppe-forest-steppe sequence. Besides replacement of forest by steppe, we found two alternative responses of vegetation to drought: shrubland expansion and savannification. The shrubland began to expand into the forest when the precipitation-brought Pacific monsoon weakened at about 4 cal ka BP. While the dry period persisted for thousands of years, we found a clear minimum in vegetation coverage during 2970 to 600 cal year BP. Keywords  Forest-steppe ecotone; LGM refugia; Vertical migration; Ecological response; Pollen; Pinus tabulaeformis; Holocene

Parts of this Thesis Have Been Published in the Following Journal Articles • Hao, Q., de Lafontaine, G., Guo, D., Gu, H., Hu, F.S., Han, Y., Song, Z., Liu, H., 2018. The critical role of local refugia in postglacial colonization of Chinese pine: joint inferences from DNA analyses, pollen records, and species distribution modeling. Ecography 41, 592–606. Reprinted from (Hao et  al., 2018). Copyright (2017), with permission from John Wiley and Sons • Hao, Q., Liu, H., Liu, X., 2016. Pollen-detected altitudinal migration of forests during the Holocene in the mountainous forest–steppe ecotone in northern China. Palaeogeography, Palaeoclimatology, Palaeoecology 446, 70–77. Reprinted from (Hao et al. 2016). Copyright (2016), with permission from Elsevier • Hao, Q., Liu, H., Yin, Y., Wang, H., Feng, M., 2014. Varied responses of forest at its distribution margin to Holocene monsoon development in northern China. Palaeogeography, Palaeoclimatology, Palaeoecology 409, 239–248. Reprinted from (Hao et al. 2014). Copyright (2014), with permission from Elsevier

ix

Acknowledgments

First and foremost, I would like to thank Prof. Hongyan Liu, who supervised me for 5 years. I have benefited tremendously from his professional knowledge, earnest academic attitude, and modesty. What he has taught me is not only the research methods but also the character of integrity and responsibility. This thesis cannot be completed without his patient instruction, valuable suggestions, and warm encouragement. Secondly, I thank Prof. Hongya Gu. She and her graduate Dongshu Guo gave me a lot of guidance during my experiments in her laboratory. With their help, the experiments were carried out smoothly. I also owe my appreciation to Prof. Feng Sheng Hu, who gave me the opportunity to study in his lab in UIUC. I would like to thank him and Dr. De Lafontaine Guillaume, who offered guidance in my data analysis, software application, and language correction. Besides, I have benefited from many other teachers. Thanks to Prof. Youxu Jiang, Shilong Piao, Zehao Shen, Yiyin Li, and Zhiheng Wang, who gave me many comments and suggestions for my thesis. Thanks to Prof. Liping Zhou and Chengjun Ji for providing the laboratory and experimental conditions. Finally, I thank all my colleagues in the laboratory for their assistance in coring the lake sediment, doing experiments, and revising this thesis. I am also grateful to all my classmates for their help during the 5 years. This work was supported by the National Natural Science Foundation of China (grant numbers 41325002, 41530747, and 31321061), the National Key Research Development Program of China (2017YFA0605101), and the 111 Project (grant number B14001).

xi

Contents

1 Introduction����������������������������������������������������������������������������������������������    1 1.1 The Impact of Climate Change on Forests ��������������������������������������     1 1.2 Importance of Forest-Steppe Ecotones to Climate Change Research������������������������������������������������������������������������������     1 1.3 Advances in Research Methods of Forest Migration������������������������     2 1.3.1 Paleoecology ������������������������������������������������������������������������     2 1.3.2 Phylogeography��������������������������������������������������������������������     3 1.3.3 Species Distribution Models ������������������������������������������������     5 1.3.4 Comprehensive Integration of Various Research Methods����������������������������������������������������������������     6 1.4 Advances in the Study of Glacial Refugia and Postglacial Forest Migrations in Northern China����������������������������������������������������������     6 1.4.1 Forest Refugia and Local Spread������������������������������������������     6 1.4.2 Vertical Distribution and Vertical Migration of Forests��������     7 1.4.3 Various Responses of Local Forests to Climate Drought ��������������������������������������������������������������     8 1.4.4 Individual Responses of Species to Climate Change������������     8 1.5 Scientific Questions and Hypotheses������������������������������������������������     9 References��������������������������������������������������������������������������������������������������    11 2 Research Area and Research Methods��������������������������������������������������   17 2.1 Overview of the Research Area��������������������������������������������������������    17 2.1.1 Climate of the Research Area ����������������������������������������������    17 2.1.2 Vegetation of the Research Area ������������������������������������������    17 2.2 Research Flow����������������������������������������������������������������������������������    19 2.3 Research Methods����������������������������������������������������������������������������    19 2.3.1 Paleoecology ������������������������������������������������������������������������    19 2.3.2 Establishment of a Regional Pollen Database����������������������    23 2.3.3 Phylogeography (Cytoplasmic DNA Analysis)��������������������    24 2.3.4 Present and LGM Species Distribution Modeling����������������    25 References��������������������������������������������������������������������������������������������������    26 xiii

xiv

Contents

3 Glacial Refugia and the Postglacial Migration of Dominant Tree Species in Northern China ��������������������������������������   31 3.1 Glacial Refugia and the Postglacial Migration of Pinus tabulaeformis����������������������������������������������������������������������    31 3.1.1 Distribution Change Based on Pollen Data Analysis������������    31 3.1.2 Genetic Diversity Within and Among P. tabulaeformis Populations������������������������������������������������    35 3.1.3 Simulated Present and LGM Distribution of P. tabulaeformis����������������������������������������������������������������    43 3.1.4 Last Glacial Maximum Distribution and Postglacial Migration of Pinus tabulaeformis����������������    45 3.2 Glacial Refugia and the Postglacial Migration of Quercus mongolica����������������������������������������������������������������������    49 3.2.1 Distribution Change Based on Pollen Data Analysis������������������������������������������������������������������������    49 3.2.2 Simulated Present and LGM Distribution of Q. mongolica��������������������������������������������������������������������    50 3.2.3 Last Glacial Maximum Distribution and Postglacial Migration of Q. mongolica��������������������������    50 3.3 Glacial Refugia and the Postglacial Migration of Betula platyphylla and Betula dahurica ��������������������������������������    51 3.3.1 Distribution Changes Based on Pollen Data Analysis����������    51 3.3.2 Simulated Present and LGM Distribution of B. platyphylla and B. dahurica ������������������������������������������  51 3.3.3 Last Glacial Maximum Distribution and Postglacial Migration of B. platyphylla and B. dahurica��������������������������������������������������������������������    52 3.4 Chapter Summary ����������������������������������������������������������������������������    52 References��������������������������������������������������������������������������������������������������    52 4 Effects of Vertical Migration on Local Vegetation��������������������������������   57 4.1 Study methods����������������������������������������������������������������������������������    57 4.1.1 Sediment Pollen Data Collection������������������������������������������    57 4.1.2 Altitude Range Calculation��������������������������������������������������    61 4.1.3 Soil Surface Pollen Records and Environmental Interpretation ����������������������������������������    62 4.2 Temporal and Spatial Change of Arboreal Pollen Percentages During the Holocene����������������������������������������������������    63 4.2.1 Temporal and Spatial Change of Pinus��������������������������������    63 4.2.2 Temporal and Spatial Change of Quercus����������������������������    65 4.2.3 Temporal and Spatial Change of Betula ������������������������������    65 4.2.4 Temporal and Spatial Change of Arboreal Pollen����������������    66 4.3 Effect of Different Factors on Soil Surface Arboreal Pollen Percentages����������������������������������������������������������������������������    66

Contents

xv

4.4 Influence of Altitude Range on Vertical Migration of Forest������������    66 4.4.1 Factors Affecting Temporal and Spatial Variations in Forest Distribution ������������������������������������������    66 4.4.2 A Conceptual Model of Forest Migration in the Mountainous Forest-Steppe Ecotone��������������������������    70 4.5 Chapter Summary ����������������������������������������������������������������������������    72 References��������������������������������������������������������������������������������������������������    72 5 Local Vegetation Dynamics and Forest Advance/Retreat��������������������   75 5.1 Overview of Huangqihai Lake and Surrounding Lakes��������������������    75 5.2 Climate and Vegetation Change of Huangqihai Lake Since 8600 Years����������������������������������������������������������������������    77 5.2.1 Sediment Coring and Chronology����������������������������������������    77 5.2.2 Regional Climate Change of Huangqihai Lake��������������������    78 5.2.3 Pollen Assemblages in Huangqihai Lake������������������������������    81 5.3 The Various Responses of Huangqihai Vegetation to Drought ��������    83 5.3.1 Holocene Climatic Change at the Limit of Monsoon Influence ����������������������������������������������������������    83 5.3.2 Response of Forest Development to Climate Drying ����������    85 5.4 Chapter Summary ����������������������������������������������������������������������������    87 References��������������������������������������������������������������������������������������������������    87 6 Forest Migration Patterns and Uncertainties����������������������������������������   91 6.1 Migration Patterns at Different Temporal and Spatial Scales ����������    91 6.1.1 Spatial Heterogeneity of Forest Migrations in Response to Climate Change��������������������������������������������    93 6.1.2 Time Lag of Forest Migrations in Response to Climate Change����������������������������������������������������������������    93 6.1.3 The Individual Responses of Different Tree Species to Climate Change ������������������������������������������    94 6.2 The Roles of Different Environmental Factors in Forest Migration ��������������������������������������������������������������������������    94 6.2.1 Dominant Roles of Climatic Factors������������������������������������    94 6.2.2 Buffering Action of Topographic Factors ����������������������������    95 6.2.3 Synergy of Disturbance Events��������������������������������������������    95 6.3 Uncertainties of this Study����������������������������������������������������������������    96 6.3.1 Uncertainties in Data Acquisition ����������������������������������������    96 6.3.2 Limitations of the Analogy Between the Past and Present Vegetation��������������������������������������������    98 References��������������������������������������������������������������������������������������������������    98 7 Main Conclusions������������������������������������������������������������������������������������  101 7.1 Long-Distance Migration and Local Diffusion of Forests����������������   101 7.2 Forest Horizontal and Vertical Migration ����������������������������������������   102 7.3 Various Responses of Local Forests to Climate Change������������������   102 7.4 The Innovations of this Book������������������������������������������������������������   103

Chapter 1

Introduction

1.1  The Impact of Climate Change on Forests Climate change has led to increasing forest mortality rates worldwide (Adams et al. 2009; Allen et al. 2010; Williams et al. 2013). In the past 100 years, the global average surface temperature has increased by approximately 0.85 °C (IPCC 2014). In the future, global warming and drought caused by greenhouse gas emissions will continue to increase (Karl and Trenberth 2003). Climate change can alter the composition, structure, and biogeographical processes of forests (Zhao and Running 2010; Peng et al. 2011) and change other environmental factors, such as pest outbreaks, habitat fragmentation, and forest fires (Dale et  al. 2001; Breshears et  al. 2005). Climate change will also lead to forest decline. Because forests account for only 30% of the global land area, they have important regulatory roles in the water cycle and biogeochemical cycle of the entire earth (Manning et  al. 2006; Bonan 2008; Anderegg et al. 2013). Declines in forest ecosystems can change the albedo and latent heat of the ground and result in feedback regulation of the climate of the earth (Rotenberg and Yakir 2010; Li et  al. 2015). Therefore, the study of forest dynamics is an important issue of global ecology.

1.2  I mportance of Forest-Steppe Ecotones to Climate Change Research The forest-steppe ecotone in northern China is located at the southeast edge of the Eurasian grasslands, from the northern section of the Greater Khingan Mountains to the Liüpan Mountains along the mountains in northern Hebei Province and the Lüliang Mountains. The forest-steppe ecotone has complex topography, vegetation types, and tree species (Liu et al. 2002a, b). This ecotone marginally influenced by © Springer Nature Singapore Pte Ltd. 2018 Q. Hao, The LGM Distribution of Dominant Tree Genera in Northern China’s Forest-steppe Ecotone and Their Postglacial Migration, Springer Theses, https://doi.org/10.1007/978-981-13-2883-1_1

1

2

1 Introduction

the East Asian Summer Monsoon (EASM) and has vegetation dynamics sensitive to monsoons (Liu et al. 2014). The climate in this region gradually became warm and wet in the early Holocene, was consistently warm and humid in the mid-Holocene, and has become dry with fluctuations since 4000 cal year BP (An et al. 2006, 2008; Herzschuh et al. 2006; Chen et al. 2008; Zhao et al. 2009). In addition, this region has shown a drought tendency over the past 50 years (Qian and Ma 2004) and a warming rate that is much higher than the global average warming rate. At the same time, precipitation has decreased in the forest-steppe ecotone of China (Liu et al. 2013). The increased temperature and decreased precipitation pose serious threats to the survival of the forest in this region. Therefore, it is of great significance to study how forests in this region have responded to past climate changes to predict future forest dynamics. The semiarid areas of northern China are located at the drought limit of the forest distribution. Afforestation in the region may lead to the poor growth of trees and even cause the degradation of grasslands and soil conditions in some areas. The implementation of a reasonable policy to convert farmland to forest or grassland areas is important to improve regional ecological conditions (He et al. 2008). “The National Forest Land Protection and Utilization Plan (2010–2020 years)” noted that the number of forest inventories will increase by 42 million hectares in 2005, with a forest coverage of over 23% by 2020. According to the present land use conditions, the land use rate of most of the southern region has reached saturation, while most of the northwest climate is unable to meet forest growth requirements. Therefore, the forest-steppe ecotone has become the main area in China for future afforestation. The study of forest dynamics in response to climate change in this region will help better guide ecological restoration efforts.

1.3  Advances in Research Methods of Forest Migration 1.3.1  Paleoecology To predict future forest dynamics, we need to understand past forest changes. However, the response of species distributions to climate change is a slow process that is usually studied with paleoecology methods (Davis and Shaw 2001), such as palynology. Palynological analysis is the most commonly used method for reconstructing paleovegetation. This method involves the determination of the characteristics of pollen genera, families, and combinations corresponding to specific vegetation types and the use of these features in the sediments as indicators of past vegetation and climates. By comparing pollen assemblages from the surface soil (corresponding to modern vegetation) and sediment, paleovegetation can be inferred based on their degree of similarity. Such studies largely rely on subjective interpretations of pollen assemblages in a single sedimentary profile; for example, decreases in tree pollen indicate that forests were replaced by grasslands.

1.3  Advances in Research Methods of Forest Migration

3

In the last 20 years, with the development of soil surface pollen research, it has been possible to establish a quantitative relationship between pollen and vegetation. However, the pollen data of surface soils are affected by many factors and may not be linearly correlated with vegetation. For regional vegetation changes, the fossil pollen map is an important tool for restoring past plant communities and studying spatial patterns of paleovegetation and paleoenvironments during specific times. Pollen percentage contours (isopolls) and isochronous lines (isochrones) are the two main forms of fossil pollen maps (Ren 1999). Ren (1999) considered a tree pollen content >40% to indicate a forest and used the pollen isoline to draw a Holocene vegetation distribution map (Ren 2007). There have been many studies on the use of thresholds of pollen from different families or genera to indicate past vegetation distribution ranges and coverages (Cao et al. 2013, 2015). At the global and regional scales, biome has become a large-scale simulation tool for the geographical distributions of vegetation, net primary productivity, and carbon balances and has become a vegetation dynamic prediction tool (Prentice and Jolly 2000; Takahara et al. 2000; Ni 2002; Ni et al. 2010). Natural vegetation can be divided into a series of biomes. Each biome is dominated by one or more plant functional types, and the climate affects the distribution of these plant functional types (Ni 2002). Using the data obtained from the palynological map published in the literature and a small amount of original data, Yu et al. (2000) used the biome method to draw a vegetation map for the Last Glacial Maximum (LGM). Although these methods have established the relationship between vegetation and pollen to some extent, there are still many problems in restoring the distribution of past vegetation. First, previous studies generally have age accuracy shortcomings. Most of the related studies analyzed previous original sediment assemblage data or digital data from the literature and then selected some points with the available age numbers and sample resolution (Zhao et al. 2009). However, the age models used by these studies vary greatly; for example, the fitting method and the “old carbon effect” will cause deviations in statistical data (Li et  al. 2006; Liu et  al. 2010). Second, the spread distance in different regions varies greatly, and we cannot accurately distinguish whether pollens are from local or long-distance areas. Thus, the pollen percentage cannot directly reflect the vegetation coverage of this area, leading to overestimated migration rates (McLachlan et al. 2005). In addition, pollen analysis cannot distinguish specific species with pollen morphology, which restricts the study of species by palynological data and represents an important disadvantage of the method (Table 1.1).

1.3.2  Phylogeography Phylogeography is a discipline related to the geographical distribution and formation processes of species (Avise et al. 1987; Wan 2010). As a branch of biogeography, phylogeography emphasizes the historical reasons for the present spatial distribution of species (Avise 1998). The change in species distributions affects the

4

1 Introduction

Table 1.1  Sources of errors associated with determining the occupancy of a species in a refugium under previous climatic periods Method Fossil records

Phylogeography

Species distribution models

Error source Uncertain fossil source (e.g., the distant propagation of spores) Fossils coming from an earlier time than the time of the study Difficult to observe (e.g., because of small populations or low reproduction rates, resulting in less fossil records) Fossil discrimination error Dating error Genetic diversity caused by hybridization, secondary contact zones, etc. Specific alleles in other study areas were not found. Genetic variation or gene drift of pioneer populations during migration Lower gene diversity or less genetic diversity, with less explanatory power Sampling is limited to existing populations. The center of gene diversity shifts over time. Restriction of dispersal (e.g., nonavailability of a potential habitat) Inappropriate model use (e.g., parameter or variable selection, mathematical expressions, incomplete input of species presence, interaction and propagation of species) Incorrect paleoclimate simulation results Low variable spatial resolution Changing niches

Modified from Gavin et al. (2014)

genetic structures of current populations (Hartl and Clark 1997); therefore, the history of the distributions of a given population can be traced back based on the genetic structures of the current plant populations (Abbott et al. 2000; Hewitt 2000; Petit et al. 2003). Quaternary glaciation (especially the LGM) has a profound impact on current biological distribution patterns, genetic structures, and evolution (Comes and Kadereit 1998; Hewitt 2000). It is of great significance to infer historical changes from phylogeography, especially for areas lacking pollen record species, and phylogeography is likely the optimal method to infer evolutionary history (Anderson et al. 2006). Pollen records are important in the reconstruction of vegetation at the community scale. However, by combining these records with phylogeography, the glacial refugia and postglacial migration processes of species can be studied. In recent years, some studies have used the genetic analyses of the leaves of natural populations obtain the genetic relationships among different populations (Chen et al. 2008; Tian et al. 2009) to make up for the inability of paleoecology research to accurately represent species-level information. Theoretically speaking, the genetic diversity of populations in LGM refugia is higher than that of young populations and often contains more unique haplotypes (Jansson and Dynesius 2002; Excoffier et al. 2009). From this point of view, we can infer the locations of the refugia of the LGM.

1.3  Advances in Research Methods of Forest Migration

5

DNA sequencing is one of the most widely used techniques in phylogeography studies (Chiang et al. 2001). The research objects are mainly concentrated in mitochondria or chloroplast DNA fragments because the genomes of these organelles have a small molecular weight, a simple structure, a single parent, and less genetic recombination, which is conducive to genetic analysis (Clegg et al. 1994). In addition, some fragments of plant cytoplasmic genes, especially introns, have a relatively high nucleotide substitution rate (Hamilton 1999), which is not only suitable for the reconstruction of phylogenetic relationships between species (Liu et  al. 2002; Wang et al. 2004) but is also suitable for the study of intraspecific phylogeographic relationships (Zhang et al. 2005; Meng and Wang 2008). However, if a population is relatively small, some haplotypes that migrated from other locations may be dominant in the migration process over a long time, and the original unique haplotype might not be obvious, resulting in lower genetic diversity and concealing the evidence of a refugium (Breen et al. 2012). Ancient DNA based on the analysis of fossils may be the most direct proof of species distributions (Parducci et al. 2012), but the experimental conditions of sediment DNA research are rigorous, and the operation is difficult; thus, this method is not widely used at present. Therefore, current studies are commonly based on the lineage patterns of natural populations.

1.3.3  Species Distribution Models Species distribution models (SDMs) are widely used in climate change biology. Based on the niche of a species, SDMs use a specific algorithm to estimate the species occurrence probability, habitat suitability, or species richness (Fang 2000). SDMs can not only be used to reflect modern distributions but also be used to reflect future or past distributions (Hu et al. 2015), with a wide range of applications in conservation biology, invasive biology, global change biology, and infectious diseases (Graham et al. 2004). At present, many species distribution models have been developed and used (e.g., GARP, MaxEnt, ENFA, Bioclim, and Domain) with different stabilities and compatibilities. According to the contrastive study of different models, MaxEnt (Maximum Entropy; Phillips et al. 2004, 2006) has a relatively good stability and prediction performance and also has simple operation advantages (Guo 2013). MaxEnt was first proposed by Jaynes (1957). The model is based on local information, infers the possible range of suitable distributions, and is considered as a maximum entropy model. MaxEnt was introduced by the AT&T Laboratory to predict the distribution of species (Phillips et al. 2004, 2006). According to the relationships between a known biological distribution area and environmental factors, MaxEnt can simulate the niche and predict the distribution of a species with a density matrix estimation (Zhao 2014). In the MaxEnt model, the number “0” means that a species does not exist, and “1” means the species exists. In this case, the simulation result becomes a continuous change between 0 and 1. MaxEnt also comes with a receiver

6

1 Introduction

operating characteristic curve (ROC) for simulation results. When the area under curve (AUC) > 0.9, the simulation results have high accuracy, and AUC > 0.85 can be accepted. In addition, the jackknife analysis can be used to obtain the importance of different influencing factors for model simulations, allowing further analysis. More importantly, MaxEnt can better address nonuniform, sparse samples with less occurrence site errors (Chen et al. 2012). Species occurrence data are typically collected by past surveys, herbarium records, and literature records, and such data are generally uneven; thus, the application of MaxEnt could solve this problem.

1.3.4  Comprehensive Integration of Various Research Methods Paleoecology, phylogeography, and species distribution models have been applied independently in the study of Quaternary climate change and species distribution reconstructions. A combination of different approaches is often required to detect small populations of plants and their postglacial migration pathways because any single approach has strengths and limitations (Table  1.1; Hu et  al. 2009; de Lafontaine et al. 2014b; Gavin et al. 2014). For example, joint inferences from fossil and genetic data provide complementary evidence about how the genetic structure of trees has emerged as a result of species shifts associated with Quaternary climatic oscillations (Magri et al. 2006; Warren et al. 2016). Species distribution modelling can complement fossil and genetic data (Dalmaris et al. 2015) by identifying potential distribution ranges at a broader spatial scale.

1.4  A  dvances in the Study of Glacial Refugia and Postglacial Forest Migrations in Northern China 1.4.1  Forest Refugia and Local Spread Studies addressing this topic traditionally relied on palaeoecological approaches (e.g., Woods and Davis 1989; Yu et al. 2000). Fossil pollen evidence indicated that many tree species in today’s boreal and temperate regions retreated to areas far south of the continental ice sheets during the Last Glacial Maximum (LGM; between 24,000 and 18,000 cal year BP) and expanded to higher latitudes during the postglacial period in Europe and North America (Huntley and Birks 1983; Ritchie 1987). In contrast, recent phylogeographic surveys across the ranges of a number of species revealed genetic structures indicative of small refugial populations of trees that survived through the LGM in northern areas near the ice sheets and sometimes even north of the ice sheets (e.g., McLachlan et al. 2005; Anderson et al. 2006; Parducci et al. 2012; Warren et al. 2016). Northern refugial populations were likely distributed discontinuously and at very low densities due to low atmospheric CO2

1.4  Advances in the Study of Glacial Refugia and Postglacial Forest Migrations…

7

concentrations and stressful climatic conditions, making it difficult to assess their past occurrences by palaeoecological reconstructions alone (de Lafontaine et  al. 2014a, b). These findings highlight the importance of local spread from these refugia in postglacial forest development and allow for more biologically realistic estimates of species migration rates for European and North American tree taxa (Feurdean et al. 2013; Snell and Cowling 2015). China was not covered by an ice sheet during the LGM (Liu 1988). Pollen-based biome reconstructions have suggested that forest ecosystems occurred south of the Yangtze River during the LGM, while steppes and deserts dominated northern China (Yu et al. 2000; Harrison et al. 2001; Ni et al. 2010). However, phylogeographical surveys suggest that temperate northern China provided glacial refugia for some tree species (Tian et al. 2009; Qiu et al. 2011). The relative roles of long-­ distance migration and northern refugia in the postglacial development of northern forests remain unclear. In addition to the mountains in northern China, the Qinghai-Tibet Plateau, as the highest, largest, and youngest plateau, is often regarded as a site of refugia. The Qinghai-Tibet Plateau was not covered by ice sheets but was covered by ice caps and valley glaciers during the glacial period (Owen 2009). The Qinghai-Tibetan Plateau also has a large surface area and many small basins, which represent advantages of the Qinghai-Tibetan Plateau as a site of glacial refugia. Many previous studies have suggested that the low-altitude edge of the Tibetan Plateau often works as a glacial refugium (Zhang et al. 2005; Yang et al. 2008), and a large number of studies have suggested that the high-altitude platform of the Qinghai-Tibet Plateau can also be considered a refugium for plant species (Wang et al. 2010; Opgenoorth et al. 2010).

1.4.2  Vertical Distribution and Vertical Migration of Forests There have been many palaeoecological studies of local vegetation dynamics in the forest-steppe ecotone in northern China (Chen et al. 2008; Jiang et al. 2006; Wang et al. 2012; Xiao et al. 2004; Yin et al. 2011, 2015). However, the inconsistency in the evidence gathered from different sediment cores in this region demonstrates that vegetation dynamics cannot be interpreted by climate conditions alone. For example, arboreal pollen percentages vary greatly among different lake sediment records even though these lakes are geographically close to each other and have similar climate conditions (Wang et al. 2012), implying that the effect of topographical factors cannot be ignored. In addition, these studies have generally focused on the south-north horizontal migration of the forest in relation to climate change (Cao et  al. 2015; Ni et  al. 2010; Yu et  al. 2000). The effects of altitudinal distribution changes on forest distribution and survival have not yet been considered. Altitude interacts with climate conditions to determine the altitudinal distribution of vegetation. Taking the forest-steppe ecotone of northern China as an example, a mosaic of forests (pine and oak forests) and steppes primarily dominates the

8

1 Introduction

lower altitudes, while birch forests are more common at the higher altitudes. It is expected that climatic drying will push the forest either to higher altitudes or southward, where there is sufficient moisture (Harsch et al. 2009). Therefore, forest vertical migration is also an important forest response to climate change. The vertical migration of forests also helps explain why the current sedimentary records are in the forest-steppe ecotone differ greatly.

1.4.3  Various Responses of Local Forests to Climate Drought Drought has a significant impact on forest reduction, but a decline in forests does not mean an increase in grassland. A forest may be replaced by grassland, savanna, or shrubland under different climatic conditions (Calvao and Palmeirim 2004; Acácio et al. 2009; Frelich and Reich 2009). Because of the strong drought resistance of shrublands (Laliberte et  al. 2004; Throop et  al. 2012), shrub coverage increases under climate drying. A decline in forest coverage is also considered a response mode to drought (Frelich and Reich 2009). To predict how forests will change under drought in the future, we need to use palynological evidence to reconstruct the past forest dynamics in the forest-steppe ecotone, especially changes in forest coverage based on pollen concentrations. The pollen sequence has mainly been used to explain the surrounding vegetation changes in the forest-steppe ecotone of northern China, especially the forest-­ grassland substitution during the Holocene (Xiao et al. 2004; Jiang et al. 2006; Chen et al. 2008; Yin et al. 2011). Although a small number of studies have taken into account changes in pollen concentrations, these studies do not associate the regional vegetation coverage based on pollen concentration with climate change. In the forest-­steppe ecotone of northern China, climate change is associated with both a long-term drought process and short, severe drought events (Zhao et al. 2009; Yin et al. 2013), which provides important research conditions for studying the responses of local vegetation to drought.

1.4.4  Individual Responses of Species to Climate Change Increasing paleoecological evidence shows that different species have different responses to environmental changes, which is known as the individual response of the species. A community is only a loose combination of species, not a whole that responds to environmental changes. Due to the differences in the life span, life history, population growth rate, and dispersal ability of different species (Liu et al. 2014), the responses of species to climate change are different. In eastern North America, Picea, Pinus, and Cyperus species flourished during 18,000~12,000  cal  year BP, while Fraxinus, Carpinus, and Populus reached a peak during 12,000~9000 cal year BP. Quercus, Tsuga, and

1.5  Scientific Questions and Hypotheses

9

Fagus species have increased since 9000 cal year BP. Peaks appeared at different times for different species, leading to changes in community species compositions. Due to research method limitations, most previous studies are at the community or genus level, and research about personalized migrating species is overlooked; however, the use of phylogeography and species distribution models has effectively solve these problems in recent years.

1.5  Scientific Questions and Hypotheses In summary, the responses of forests to climate change demonstrate different patterns at different spatiotemporal scales. Previous studies have mainly focused on forest refugia and postglacial rapid migration from southern refugia to the north. Although northern China was not covered by a glacial sheet during the LGM, it is generally believed that the forest moved to the south of the Yangtze River during the LGM (Yu et al. 2000). Were there any refugia for forests in northern China during the LGM? How has the forest migrated since the LGM? Answering these questions is of great significance for predicting how forests in northern China (especially the forest in forest-steppe ecotone) respond to future climate change. According to different temporal and spatial scales, we choose the following three interrelated scientific questions: 1. Where were the refugia of the dominant trees in the LGM in northern China? How did these trees migrate after the LGM? This book focuses on the study of Pinus tabulaeformis. Chinese pine is a widespread coniferous species endemic to northern China. The northern limit of the P. tabulaeformis distribution matches the northern margin of the monsoon (Wu 1980), located around the 400-mm isohyet. The core of the current range of P. tabulaeformis is in mountainous areas surrounding the Loess Plateau of northern China (Qilian, Qinling, and Taihang Mountains). Chen et al. (2008) investigated the phylogeography of P. tabulaeformis and proposed three low-elevation glacial macrorefugia by the Yellow River, near the Yellow Sea, and within the Sichuan Basin. Thus, according to these authors, Chinese pine shifted southward within and beyond the southern part of the species’ current range during the last glacial period and expanded northward during the postglacial period, as proposed for coniferous forests (Winkler and Wang 1993; Wang and Sun 1994; Yu et al. 2000; Harrison et al. 2001). However, environmental heterogeneity in mountainous areas often provides microhabitats allowing plant species to survive locally through periods of fluctuating climatic conditions (i.e., microrefugia) (Qiu et  al. 2011; Papageorgiou et  al. 2014; Birks 2015). It is possible that the complex topography of northern China allowed the local persistence of tree populations. In addition, the Otindag Sandy Land, located at the northern limit of the P. tabulaeformis distribution, has a high groundwater level, and P. tabulaeformis is sparsely distributed in this region even though the macroclimate is unsuitable for the species. Locally elevated groundwater

10

1 Introduction

levels might also have supplied enough water for P. tabulaeformis during the LGM. Considering the strong drought resistance of P. tabulaeformis, we proposed hypothesis 1: this species persisted through the dry climatic period of the LGM in multiple refugia within topographically complex areas north of the Yangtze River, i.e., the Qilian, Qinling, and Taihang Mountains surrounding the Loess Plateau of northern China, as well as the Otindag Sandy Land. As dominant tree species in northern China, Betula and Quercus species are also widespread in mid and low mountains, except for Pinus tabulaeformis (Liu et al. 2002a). This paper also used the pollen and species distribution model (MaxEnt) to analyze the possible refugia and the distribution process during the LGM. Based on these studies, we can better understand the current vegetation patterns of different trees with different ecological habits. 2. How did topographic patterns affect forest dynamics after the glacial period in northern China? Just as mentioned above, altitude interacts with climate conditions to determine the altitudinal distribution of vegetation. It is expected that climatic drying will push the forest either to higher altitudes or southward, areas with sufficient moisture (Harsch et  al. 2009). However, the successful altitudinal migration of vegetation depends on the altitudinal range. Thus, our 2nd hypothesis is as follows: the arboreal pollen fraction in a watershed is larger if the altitudinal range around a lake is high because higher altitude ranges can offer more different types of habitats for forest survival and more possibilities for the altitudinal migration of forests. 3. How did forest types and coverage change in the forest-steppe ecotone during the Holocene? There are two main vegetation patterns in the forest-steppe ecotone of northern China. On the one hand, the forest shows a patchy distribution in steep and shady slopes, and steppes exist in sunny and gentle slopes. On the other hand, steppes are dominated with Ulmus pumila and Caragana microphylla in the sandy land (Liu et  al. 2002a). In this area, climate affects the distribution of vegetation mainly through soil moisture. Accordingly, the third hypothesis put forth in this paper is as follows: there are various ways a forest responds to climate. There may be changes in forest patch size and sparsity and the replacement of forests by shrubs and grasslands. The occurrence of different patterns may be driven by drought duration and drought intensity. The verification of this hypothesis is mainly achieved through a detailed analysis of the high-resolution sedimentary section, which can provide a basis for understanding the refugia and migration of forests at larger spatial scales.

References

11

References Abbott, R.J., Smith, L.C., Milne, R.I., Crawford, R.M., Wolff, K., Balfour, J.  2000. Molecular analysis of plant migration and refugia in the Arctic. Science, 289, 1343–1346. Acácio, V., Holmgren, M., Rego, F., Moreira, F., Mohren, G.M. 2009. Are drought and wildfires turning Mediterranean cork oak forests into persistent shrublands? Agroforestry Systems, 76, 389–400. Adams, H.D., Guardiola-Claramonte, M., Barron-Gafford, G.A., Villegas, J.C., Breshears, D.D., Zou, C.B., Troch, P.A., Huxman, T.E. 2009. Temperature sensitivity of drought-induced tree mortality portends increased regional die-off under global-change-type drought. Proceedings of the National Academy of Sciences, 106, 7063–7066. Allen, C.D., Macalady, A.K., Chenchouni, H., Bachelet, D., McDowell, N., Vennetier, M., Kitzberger, T., Rigling, A., Breshears, D.D., Hogg, E. 2010. A global overview of drought and heat-induced tree mortality reveals emerging climate change risks for forests. Forest Ecology and Management, 259, 660–684. An, C., Chen, F., Barton, L. 2008. Holocene environmental changes in Mongolia: a review. Global and Planetary Change, 63, 283–289. An, C., Feng, Z., Barton, L. 2006. Dry or humid? Mid-Holocene humidity changes in arid and semi-arid China. Quaternary Science Reviews, 25, 351–361. Anderegg, W.R., Kane, J.M., Anderegg, L.D. 2013. Consequences of widespread tree mortality triggered by drought and temperature stress. Nature Climate Change, 3, 30–36. Anderson, L.L., Hu, F.S., Nelson, D.M., Petit, R.J., Paige, K.N. 2006. Ice-age endurance: DNA evidence of a white spruce refugium in Alaska. Proceedings of the National Academy of Sciences, 103, 12447–12450. Avise, J.C. 1998. The history and purview of phylogeography: a personal reflection. Molecular Ecology, 7, 371–379. Avise, J.C., Arnold, J., Ball, R.M., Bermingham, E., Lamb, T., Neigel, J.E., Reeb, C.A. and Saunders, N.C. 1987. Intraspecific Phylogeography: the Mitochondrial DNA Bridge between Population Genetics and Systematics. Annual Review of Ecology and Systematics, 18, 489–522. Birks, H.J.B. 2015. Some reflections on the refugium concept and its terminology in historical biogeography, contemporary ecology and global-change biology. Biodiversity, 16, 196–212. Bonan, G.B. 2008. Forests and climate change: forcings, feedbacks, and the climate benefits of forests. Science, 320, 1444–1449. Breen, A.L., Murray, D.F., Olson, M.S. 2012. Genetic consequences of glacial survival: the late Quaternary history of balsam poplar (Populus balsamifera L.) in North America. Journal of Biogeography, 39, 918–928. Breshears, D.D., Cobb, N.S., Rich, P.M., Price, K.P., Allen, C.D., Balice, R.G., Romme, W.H., Kastens, J.H., Floyd, M.L., Belnap, J. 2005. Regional vegetation die-off in response to global-­ change-­type drought. Proceedings of the National Academy of Sciences of the United States of America, 102, 15144–15148. Calvao, T., Palmeirim, J.  2004. Mapping Mediterranean scrub with satellite imagery: biomass estimation and spectral behaviour. International Journal of Remote Sensing, 25, 3113–3126. Cao, X., Herzschuh, U., Ni, J., Zhao, Y., Böhmer, T. 2015. Spatial and temporal distributions of major tree taxa in eastern continental Asia during the last 22,000 years. The Holocene, 25, 79–91. Cao, X., Ni, J., Herzschuh, U., Wang, Y., Zhao, Y. 2013. A late Quaternary pollen dataset from eastern continental Asia for vegetation and climate reconstructions: Set up and evaluation. Review of Palaeobotany and Palynology, 194, 21–37. Chen, F., Yu, Z., Yang, M., Ito, E., Wang, S., Madsen, D.B., Huang, X., Zhao, Y., Sato, T., John B Birks, H. 2008. Holocene moisture evolution in arid central Asia and its out-of-phase relationship with Asian monsoon history. Quaternary Science Reviews, 27, 351–364.

12

1 Introduction

Chen, X., Lei, Y., Zhang, X., Jia, H. 2012. Effects of sample sizes on accuracy and stability of maximum entropy model in predicting species distribution. Scientia Silvae Sinicae, 48, 53–59. (in Chinese with English abstract) Chiang, S.H., Baumann, C.A., Kanzaki, M., Thurmond, D.C., Neudauer, C.L., Macara, I.G., Pessin, J.E., Saltiel, A.R. 2001. Insulin-stimulated GLUT4 translocation requires the CAP-­ dependent activation of TC10. Nature, 410, 944–948. Clegg, M.T., Gaut, B.S., Learn, G.H., Monton, B.R. 1994. Rates and patterns of chloroplast DNA evolution. Proceedings of the National Academy of Sciences, 91, 6795–6801. Comes, H.P., Kadereit, J.W. 1998. The effect of Quaternary climatic changes on plant distribution and evolution. Trends in Plant Science, 3, 432–438. Dale, V.H., Joyce, L.A., McNulty, S., Neilson, R.P., Ayres, M.P., Flannigan, M.D., Hanson, P.J., Irland, L.C., Lugo, A.E., Peterson, C.J. 2001. Climate Change and Forest Disturbances Climate change can affect forests by altering the frequency, intensity, duration, and timing of fire, drought, introduced species, insect and pathogen outbreaks, hurricanes, windstorms, ice storms, or landslides. BioScience, 51, 723–734. Dalmaris, E., Ramalho, C.E., Poot, P., Veneklaas, E.J., Byrne, M. 2015. A climate change context for the decline of a foundation tree species in south-western Australia: insights from phylogeography and species distribution modelling. Annals of botany, mcv044. Davis, M.B., Shaw, R.G. 2001. Range shifts and adaptive responses to Quaternary climate change. Science, 292, 673–679. de Lafontaine, G., Amasifuen Guerra, C.A., Ducousso, A., Petit, R.J. 2014a. Cryptic no more: soil macrofossils uncover Pleistocene forest microrefugia within a periglacial desert. New Phytologist, 204, 715–729. de Lafontaine, G., Amasifuen Guerra, C.A., Ducousso, A., Sánchez-Goñi, M.F., Petit, R.J. 2014b. Beyond skepticism: inferring cryptic refugia using multiple lines of evidence. New Phytologist, 204, 450–454. Excoffier, L., Foll, M., Petit, R.J. 2009. Genetic consequences of range expansions. Annual Review of Ecology, Evolution, and Systematics, 40, 481–501. Fang, J.  2000. Global Ecology: Climate Change and Ecological Response. Beijing: Higher Education Press. Feurdean, A., Bhagwat, S.A., Willis, K.J., Birks, H.J.B., Lischke, H., Hickler, T. 2013. Tree Migration-Rates: Narrowing the Gap between Inferred Post-Glacial Rates and Projected Rates. Plos one, 8, e71797. Frelich, L.E., Reich, P.B. 2009. Will environmental changes reinforce the impact of global warming on the prairie-forest border of central North America? Frontiers in Ecology and the Environment, 8, 371–378. Guo, N. 2013. Potential distribution prediction of medium flavor type of flue-cured tobacco. Thesis for master degree of Southwest University. Gavin, D.G., Fitzpatrick, M.C., Gugger, P.F., Heath, K.D., Rodríguez-Sánchez, F., Dobrowski, S.Z., Hampe, A., Hu, F.S., Ashcroft, M.B., Bartlein, P.J. 2014. Climate refugia: joint inference from fossil records, species distribution models and phylogeography. New Phytologist, 204, 37–54. Graham, C.H., Ron, S.R., Santos, J.C., Schneider, C.J., Moritz, C. 2004. Integrating phylogenetics and environmental niche models to explore speciation mechanisms in dendrobatid frogs. Evolution 58, 1781–1793. Hamilton, M.B. 1999. Four primer pairs for the amplification of chloroplast intergenic regions with intraspecific variation. Molecular Ecology, 8, 521–522. Harrison, S., Yu, G., Takahara, H., Prentice, I. 2001. Palaeovegetation Communications arising: diversity of temperate plants in east Asia. Nature, 413, 129–130. Harsch, M.A., Hulme, P.E., McGlone, M.S., Duncan, R.P. 2009. Are treelines advancing? A global meta-analysis of treeline response to climate warming. Ecology Letters, 12, 1040–1049. Hartl, D.L., Clark, A.G. 1997. Principles of population genetics. Sinauer Associates, Inc. Publishers, Sunderland, UK.

References

13

He, S., Liu, H., Ren, J., Yin, Y. 2008. Landform-climate-vegetation patterns and countermeasures for vegetation rehabilitation of forest-steppe ecotone on southeastern Inner Mongolia Plateau. Scientia Geographica Sinica, 28, 253–258. (in Chinese with English abstract) Herzschuh, U., Winter, K., Wünnemann, B., Li, S. 2006. A general cooling trend on the central Tibetan Plateau throughout the Holocene recorded by the Lake Zigetang pollen spectra. Quaternary International, 154, 113–121. Hewitt, G. 2000. The genetic legacy of the Quaternary ice ages. Nature, 405, 907–913. Hu, F.S., Hampe, A., Petit, R.J. 2009. Paleoecology meets genetics: deciphering past vegetational dynamics. Frontiers in Ecology and the Environment, 7, 371–379. Hu, Z., Zhang, Y., Yu, H. 2015. Simulation of Stipa purpurea distribution pattern on Tibetan Plateau based on Maxent model and GIS. Chinese Journal of Applied Ecology, 26, 505–511. (in Chinese with English abstract) Huntley, B., Birks, H.J.B. 1983. An atlas of past and present pollen maps for Europe: 0–13000 years ago. Cambridge University Press, Cambridge, UK. IPCC, 2014. Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Jansson, R., Dynesius, M. 2002. The fate of clades in a world of recurrent climatic change: Milankovitch oscillations and evolution. Annual Review of Ecology and Systematics, 33, 741–777. Jaynes, E.T. 1957. Information theory and statistical mechanics. Physical Review, 106, 620. Jiang, W., Guo, Z., Sun, X., Wu, H., Chu, G., Yuan, B., Hatté, C., Guiot, J. 2006. Reconstruction of climate and vegetation changes of Lake Bayanchagan (Inner Mongolia): Holocene variability of the East Asian monsoon. Quaternary Research, 65, 411–420. Karl, T.R., Trenberth, K.E. 2003. Modern global climate change. Science, 302, 1719–1723. Laliberte, A.S., Rango, A., Havstad, K.M., Paris, J.F., Beck, R.F., McNeely, R., Gonzalez, A.L. 2004. Object-oriented image analysis for mapping shrub encroachment from 1937 to 2003 in southern New Mexico. Remote Sensing of Environment, 93, 198–210. Li, Y., Zhao, M., Motesharrei, S., Mu, Q., Kalnay, E., Li, S. 2015. Local cooling and warming effects of forests based on satellite observations. Nature Communication, 6, 6603. Li, Y., Willis, K., Zhou, L., Cui, H. 2006. The impact of ancient civilization on the northeastern Chinese landscape: palaeoecological evidence from the Western Liaohe River Basin, Inner Mongolia. The Holocene, 16, 1109–1121. Liu, H., Cui, H., Yu, P., Huang, Y. 2002a. The origin of remnant forest stands of Pinus tabulaeformis in southeastern Inner Mongolia. Plant Ecology, 158, 139–151. Liu, H., Xu, L., Cui, H. 2002b. Holocene history of desertification along the woodland-steppe border in northern China. Quaternary Research, 57, 259–270. Liu, H., Yin, Y., Hao, Q., Liu, G. 2014. Sensitivity of temperate vegetation to Holocene development of East Asian monsoon. Quaternary Science Reviews, 98, 126–134. Liu, H., Park Williams, A., Allen, C.D., Guo, D., Wu, X., Anenkhonov, O.A., Liang, E., Sandanov, D.V., Yin, Y., Qi, Z. 2013. Rapid warming accelerates tree growth decline in semi-arid forests of Inner Asia. Global Change Biology, 19, 2500–2510. Liu, H., Yin, Y., Zhu, J., Zhao, F., Wang, H. 2010. How did the forest respond to Holocene climate drying at the forest-steppe ecotone in northern China? Quaternary International, 227, 46–52. Liu, J., Gao, T., Chen, Z., Lu, A. 2002. Molecular phylogenyand biogeography of the Qinghai-­ Tibet Plateau endemic Nannoglottis (Asteraceae). Molecular Phylogenetics and Evolution, 23, 307–325. Liu, K. 1988. Quaternary history of the temperate forests of China. Quaternary Science Reviews, 7, 1–20. Magri, D., Vendramin, G.G., Comps, B., Dupanloup, I., Geburek, T., Gömöry, D., Latałowa, M., Litt, T., Paule, L., Roure, J.M. 2006. A new scenario for the Quaternary history of European

14

1 Introduction

beech populations: palaeobotanical evidence and genetic consequences. New Phytologist, 171, 199–221. Manning, A.D., Fischer, J., Lindenmayer, D.B. 2006. Scattered trees are keystone structures– implications for conservation. Biological Conservation, 132, 311–321. McLachlan, J.S., Clark, J.S., Manos, P.S. 2005. Molecular indicators of tree migration capacity under rapid climate change. Ecology, 86, 2088–2098. Meng, L., Wang, Y. 2008. Phylogeography of Hippophae neurocarpa (Elaeagnaceae) inferred from the chloroplast DNA trnL-F sequence variation. Journal of Systematics and Evolution 46(1), 32–40. (in Chinese with English abstract) Ni, J.  2002. Biome models: main principles and applications. Acta Phytoecologica Sinica, 26, 481–488. (in Chinese with English abstract) Ni, J., Yu, G., Harrison, S.P., Prentice, I.C. 2010. Palaeovegetation in China during the late Quaternary: Biome reconstructions based on a global scheme of plant functional types. Palaeogeography, Palaeoclimatology, Palaeoecology, 289, 44–61. Opgenoorth, L., Vendramin, G.G., Mao, K., Miehe, G., Miehe, S., Liepelt, S., Liu, J., Ziegenhagen, B. 2010. Tree endurance on the Tibetan Plateau marks the world’s highest known tree line of the Last Glacial Maximum. New Phytologist, 185, 332–342. Owen, L.A. 2009. Latest Pleistocene and Holocene glacier fluctuations in the Himalaya and Tibet. Quaternary Science Reviews, 28, 2150–2164. Papageorgiou, A.C., Tsiripidis, I., Mouratidis, T., Hatziskakis, S., Gailing, O., Eliades, N.G.H., Vidalis, A., Drouzas, A.D., Finkeldey, R. 2014. Complex fine-scale phylogeographical patterns in a putative refugial region for Fagus sylvatica (Fagaceae). Botanical Journal of the Linnean Society, 174, 516–528. Parducci, L., Jørgensen, T., Tollefsrud, M.M., Elverland, E., Alm, T., Fontana, S.L., Bennett, K.D., Haile, J., Matetovici, I., Suyama, Y. 2012. Glacial Survival of Boreal trees in northern Scandinavia. Science, 335, 1083–1086. Peng, C., Ma, Z., Lei, X., Zhu, Q., Chen, H., Wang, W., Liu, S., Li, W., Fang, X., Zhou, X. 2011. A drought-induced pervasive increase in tree mortality across Canada's boreal forests. Nature Climate Change, 1, 467–471. Petit, R.J., Aguinagalde, I., de Beaulieu, J.L., Bittkau, C., Brewer, S., Cheddadi, R., Ennos, R., Fineschi, S., Grivet, D., Lascoux, M. 2003. Glacial refugia: hotspots but not melting pots of genetic diversity. Science, 300, 1563–1565. Phillips, S. J., Dudík, M., Schapire, R.E. 2004. A maximum entropy approach to species distribution modeling. In: Proceedings of the Twenty-First International Conference on Machine Learning. ACM Press, New York, USA. Phillips, S.J., Anderson, R.P., Schapire, R.E. 2006. Maximum entropy modeling of species geographic distributions. Ecological Modelling, 190, 231–259. Prentice, I.C., Jolly, D. 2000. Mid-Holocene and glacial-maximum vegetation geography of the northern continents and Africa. Journal of Biogeography, 27, 507–519. Qian, W., Ma, Z. 2004. Trend analysis and prediction research on aridification of the northern China. Beijing: China Meteorological Press. Qiu, Y., Fu, C., Comes, H.P. 2011. Plant molecular phylogeography in China and adjacent regions: tracing the genetic imprints of Quaternary climate and environmental change in the world’s most diverse temperate flora. Molecular Phylogenetics and Evolution, 59, 225–244. Ren, G. 1999. Mapping and analyzing the Holocene pollen data of northeast china. Acta Palaeontologica Sinica, 38, 365–385. (in Chinese with English abstract) Ren, G., 2007. Changes in forest cover in China during the Holocene. Vegetation History and Archaeobotany, 16, 119–126. Ritchie, J. C. 1987. Postglacial vegetation of Canada. Cambridge University Press. Rotenberg, E., Yakir, D. 2010. Contribution of semi-arid forests to the climate system. Science, 327, 451–454.

References

15

Snell, R.S., Cowling, S.A. 2015. Consideration of dispersal processes and northern refugia can improve our understanding of past plant migration rates in North America. Journal of Biogeography, 42, 1677–1688. Takahara, H., Sugita, S., Harrison, S.P., Miyoshi, N., Morita, Y., Uchiyama, T. 2000. Pollen-based reconstructions of Japanese biomes at 0, 6000 and 18,000 14C yr bp. Journal of Biogeography, 27, 665–683. Throop, H.L., Reichmann, L.G., Sala, O.E., Archer, S.R. 2012. Response of dominant grass and shrub species to water manipulation: an ecophysiological basis for shrub invasion in a Chihuahuan Desert Grassland. Oecologia, 169, 373–383. Tian, B., Liu, R., Wang, L., Qiu, Q., Chen, K., Liu, J. 2009. Phylogeographic analyses suggest that a deciduous species (Ostryopsis davidiana Decne., Betulaceae) survived in northern China during the Last Glacial Maximum. Journal of Biogeography, 36, 2148–2155. Wan, Q. 2010. Phylogeographical Study on Chrysanthemumzawadskii species complex. Thesis for PhD’s Degree of Peking University. Wang, H., Liu, H., Zhao, F., Yin, Y., Zhu, J., Snowball, I. 2012. Early- and mid-Holocene palaeoenvironments as revealed by mineral magnetic, geochemical and palynological data of sediments from Bai Nuur and Ulan Nuur, southeastern inner Mongolia Plateau, China. Quaternary International, 250, 100–118. Wang, H., Qiong, L., Sun, K., Lu, F., Wang, Y., Song, Z., Wu, Q., Chen, J., Zhang, W. 2010. Phylogeographic structure of Hippophae tibetana (Elaeagnaceae) highlights the highest microrefugia and the rapid uplift of the Qinghai-Tibetan Plateau. Molecular Ecology, 19, 2964–2979. Wang, P., Sun, X. 1994. Last Glacial Maximum in China: comparison between land and sea. Catena 23, 341–353. Wang, Y., Li, X., Hao, G., Liu, J.  2004. Molecular phylogeny and biogeography of Androsace (Primulaceae) and the convergent evolution of cushion morphology. Acta Phytotaxonomica Sinica, 6, 481–499. (in Chinese with English abstract) Warren, E., de Lafontaine, G., Gérardi, S., Senneville, S., Beaulieu, J., Perron, M., Jaramillo-­ Correa, J.P., Bousquet, J. 2016. Joint inferences from cytoplasmic DNA and fossil data provide evidence for glacial vicariance and contrasted postglacial dynamics in tamarack, a transcontinental conifer. Journal of Biogeography, doi:https://doi.org/10.1111/jbi.12675. Winkler, M.  G., Wang, P.  K. 1993. The Late Quaternary vegetation and climate of China. In: Wright, H. E. (ed.), Global climates since the Last Glacial Maximum. Univ. Minnesota Press, pp. 221–264. Williams, A., Allen, C.D., Macalady, A.K., et al. 2013. Temperature as a potent driver of regional forest drought stress and tree mortality. Nature Climate Change, 3, 292–297. Woods, K. D., Davis, M. B. 1989. Paleoecology of range limits: beech in the Upper Peninsula of Michigan. Ecology, 70, 681–696. Wu, Z. 1980. The vegetation of China. Beijing: Science Press. Xiao, J., Xu, Q., Nakamura, T., Yang, X., Liang, W., Inouchi, Y. 2004. Holocene vegetation variation in the Daihai Lake region of north-central China: a direct indication of the Asian monsoon climatic history. Quaternary Science Reviews, 23, 1669–1679. Yang, F., Li, Y., Ding, X. Wang, X. 2008. Extensive population expansion of Pedicularis longiflora (Orobanchaceae) on the Qinghai-Tibetan Plateau and its correlation with the Quaternary climate change. Molecular Ecology, 17, 5135–5145. Yin, Y., Liu, H., Hao, Q. 2015. The role of fire in the late Holocene forest decline in semi-arid North China. The Holocene, 26, 93–101. Yin, Y., Liu, H., Liu, G., Hao, Q., Wang, H., 2013. Vegetation responses to mid-Holocene extreme drought events and subsequent long-term drought on the southeastern Inner Mongolian Plateau, China. Agricultural and forest meteorology, 178, 3–9. Yin, Y., Liu, H., He, S., Zhao, F., Zhu, J., Wang, H., Liu, G., Wu, X. 2011. Patterns of local and regional grain size distribution and their application to Holocene climate reconstruction in semi-arid Inner Mongolia, China. Palaeogeography, Palaeoclimatology, Palaeoecology, 307, 168–176.

16

1 Introduction

Yu, G., Chen, X., Ni, J., Cheddadi, R., Guiot, J., Han, H., Harrison, S.P., Huang, C., Ke, M., Kong, Z. 2000. Palaeovegetation of China: a pollen data-based synthesis for the mid-Holocene and Last Glacial Maximum. Journal of Biogeography, 27, 635–664. Zhang, Q., Chiang, T.Y., George, M., Liu, J., Abbott, R.J. 2005. Phylogeography of the Qinghai-­ Tibetan Plateau endemic Juniperus przewalskii (Cupressaceae) inferred fromchloroplast DNA sequence variation. Molecular Ecology, 14, 3513–3524. Zhao, J. 2014. Study on Habitat Suitability of Grape Phylloxera (Daktulosphaira vitifoliae Fitch) in China. Thesis for Master’s Degree of Northwest A&F University. Zhao, M., Running, S. 2010. Drought-induced reduction in global terrestrial net primary production from 2000 through 2009. Science, 329, 940–943. Zhao, Y., Yu, Z., Chen, F. 2009. Spatial and temporal patterns of Holocene vegetation and climate changes in arid and semi-arid China. Quaternary International, 194, 6–18.

Chapter 2

Research Area and Research Methods

2.1  Overview of the Research Area 2.1.1  Climate of the Research Area The study area is mainly located in the forest-steppe ecotone in northern China (Fig. 2.1) and in the marginal area of the East Asian monsoon influence. With the gradually weakening influence of the East Asian monsoon from the southeast to northwest, the temperature and precipitation in the region decrease from the southeast to northwest (Hou 1988; Zhou 1992; Qin 2005). The mean annual temperature (MAT) in the region is −2~12  °C, and the mean annual precipitation (MAP) is 150~650  mm. The climate is mainly controlled by the Siberian-Mongolian high-­ pressure system in winter, resulting in cold and dry environmental conditions. Warm temperatures and high humidity are caused by the prevailing Asian monsoon in summer. Variations in the soil water content have obvious seasonal changes, and the soil is usually dry from October to April and wet from May to September (Yin 2012).

2.1.2  Vegetation of the Research Area To reflect the glacial refugia and postglacial migration of major tree species in the forest-steppe ecotone in northern China, this study also extended to the east, i.e., to 95°~130°E, 30°~50°N, encompassing this region in addition to the typical profiles located in the forest-steppe ecotone. The present book also collected the Quaternary palynological literature in this area, which includes the temperate steppe and warm temperate deciduous broad-leaved forests and involves several major surrounding vegetation zones, including subtropical evergreen broad-leaved forest, temperate coniferous forest, temperate desert, and alpine vegetation area (Fig. 2.1). © Springer Nature Singapore Pte Ltd. 2018 Q. Hao, The LGM Distribution of Dominant Tree Genera in Northern China’s Forest-steppe Ecotone and Their Postglacial Migration, Springer Theses, https://doi.org/10.1007/978-981-13-2883-1_2

17

18

2  Research Area and Research Methods

Fig. 2.1  Study area (black rectangle), the forest-steppe ecotone (orange curve), sediment sites (black dots), and the location of Huangqihai

Current vegetation types and patterns are similar in the areas around all eight lakes, with forest patches embedded into the widely distributed temperate steppe. On the surrounding hills and mountains, three types of tree genera, Pinus, Quercus, and Betula, are dominant (Wu 1980). In our study area, Pinus tabulaeformis is the main representative of its genus. P. tabulaeformis is an endemic pine species in northern China. The northern edge of its distribution corresponds to the northern margin of the Asian monsoon influence, and there have been several detailed studies on this species (Chen et al. 2008; Liang and Eckstein 2006; Liu et al. 2009; Shi et al. 2008; Xu 1990). The genus Quercus is distributed widely in China and is also an important component in many forest ecosystems with coniferous and broad-leaved tree species. Quercus mongolica is the main species of this genus in the mountains in our study areas. Quercus mongolica live in warm and humid climate areas but also have the ability to tolerate lower levels of cold and drought. Betula platyphylla and Betula davurica are the major broad-leaved tree species in northern China. They are widely distributed in eastern Asia and often play a pioneering role in secondary forests. These two species both favor light and are adapted to steep terrains. As the dominant species in mid-low mountain forests, Betula platyphylla is distributed in mountainous areas with relatively high altitudes, while Betula davurica is adapted to relatively low altitudes. By comparison, we can summarize the different responses of these two species, which are also the dominant tree species in forest-steppe ecotones (Liu et  al. 2002a); thus, research on migration processes can help us to better understand vegetation dynamics under climate change in the current forest-grass ecotone.

2.3  Research Methods

19

Fig. 2.2  Research flow chart

2.2  Research Flow This book mainly studies forest migration processes and dynamic responses to climate change. According to the study area scale, the book is divided into three parts: (1) the LGM refugia and postglacial horizontal migration of three dominant tree genera in the forest-steppe ecotone in China (see Chap. 3), (2) the vertical migration of forests in the southeast margin of the Inner Mongolia Plateau and the effect of topographic factors on this migration (see Chap. 4), and (3) the local response of vegetation evolution to climate change at Huangqihai Lake in the forest-steppe ecotone (see Chap. 5) (Fig. 2.2).

2.3  Research Methods 2.3.1  Paleoecology All samples collected in the field were dried in the laboratory for the follow-up experiments. In this paper, we mainly use paleoecology methods, including AMS 14 C dating, palynological analysis, and physical and chemical property analysis. 2.3.1.1  AMS 14C Dating For every lake sediment or profile, we observed the changes in the sedimentary facies and selected samples from the surface, bottom, and turning points of the sedimentary facies to conduct the AMS 14C dating. This dating method has a high

20

2  Research Area and Research Methods

accuracy and fast measurement speed and requires small sample amounts. The measurements were completed in the AMS laboratory of Peking University. IntCal13 was used to correct the 14C age results (Stuiver et al. 1998; Reimer et al. 2013), and we used the corrected ages (Cal year BP) in this book. 2.3.1.2  Palynological Analysis For each sample of 4–8 cm3 sediments, 30% hydrochloric acid (HCl) was added. The sample was left for 18 h to remove carbonates and then boiled in 10% sodium hydroxide (NaOH) for 30 min to remove organic matter. Next, the samples were sieved though 180  μm and 7  μm sieves and treated with a heavy liquid solution (KI + HI + Zn) with a specific gravity of approximately 2.0 to extract pollen from the sediments (http://www.ucl.ac.uk/GeolSci/micropal/spore.html). Finally, the extracted samples were treated with acetolysis (Moore et al. 1991). A tablet of Lycopodium spores (approximately 27,637 grains) was added to each of the samples as exotic markers; thus, the pollen concentration could be obtained (Maher 1981). Pollen taxa were identified under an Olympus optical microscope at 400× magnification, and more than 250 terrestrial pollen grains were counted for each sample (Liu et al. 2010). The percentage of pollen belonging to one kind of family or genus was accounted for (excluding aquatic plants and spores) according to the following equation:

A = n / N × 100%

(2.1)

where A: the percentage of a certain type of pollen in a sample n: the number of grains of a certain type of pollen N: the total number of terrestrial plant pollen grains counted in a sample The absolute concentration was calculated as follows:

Pc = L / M × ( N / S )



(2.2)

where Pc: the concentration of a certain type of pollen or the total pollen concentration (grain/cm3) L: the number of Lycopodium grains added in the sample M: the number of Lycopodium grains counted in the sample N: the number of grains belonging to a certain type of pollen in the sample S: the volume of the sample In this paper, the palynological assemblages in the sediments reflect vegetation changes in the past, and the concentrations of different spores or pollen can reflect the vegetation coverage. The pollen percentages of the main taxa, the ratio of Artemisia to Chenopodiaceae (A/C), the ratio of arboreal pollen to non-arboreal pollen (AP/NAP), and the pollen

2.3  Research Methods

21

concentrations of all taxa are presented. The A/C ratio has been used as a good indicator of available moisture in arid and semiarid regions because Artemisia and Chenopodiaceae indicate steppe- and desert-dominated vegetation, respectively (Herzschuh 2007; Huang et al. 2009; Zhao et al. 2007; Zhao and Yu 2012). Recent work has shown that the A/C ratio has a positive relationship with moisture in areas with a MAP less than 450–500 mm (Zhao and Yu 2012). AP/NAP is better than A/C for indicating the available moisture in forest regions with a MAP higher than 450–500 mm, with higher values of AP/NAP indicating higher available moisture (An et al. 2006). 2.3.1.3  Grain Size Approximately 0.3–0.4 g of each sample was used for grain size measurements with a Malvern Mastersizer 2000. Before the measurement, 10–20 ml of 30% H2O2 was added until no more bubbles formed to remove organic matter. Then, the samples were boiled with enough 10% HCl to remove carbonates. Deionized water was used to rinse acidic ions until the pH value was close to 7. Finally, samples were boiled with 10 ml 0.5 mol/L (NaPO3)6. Within 3 h, these samples were treated on an ultrasonic vibrator for 30 s, and the grain distribution was divided into three classes: clay ( 63 μm). In general, the sand percentage indicated the wind strength in the arid region (De Deckker et al. 1991; Porter and An 1995; Qiang et al. 2007) and the hydrological activity in the humid region (Wu et al. 2006). In the semiarid region, the sand percentage represented both the wind and water activity intensity. Because the precipitation was relatively low and the evaporation was high, most precipitation evaporated, resulting in weak runoff. At the same time, because of the small difference in elevation across the basin and the disc shape of the lake, it was difficult for runoff to carry sand to the lake center directly. Additionally, the intense wind activity, poor vegetation coverage, and long freeze-up period led to coarse particles being moved into the center of the lake by the wind (Zhai et al. 2006). Therefore, the sand carried to the lake center was mostly transported by wind rather than by water; thus, the percentage of coarse particles indicated the winter wind activity in the dry and cold climate (Yin et al. 2011). 2.3.1.4  Total Organic Carbon (TOC) and Total Nitrogen (TN) After removing the roots and other macro-organic remains and grinding the samples to pass through a 149 μm soil sieve, total carbon (TC) and total nitrogen (TN) were measured by an Elementar Vario EL (Germany). To obtain the total organic carbon (TOC) values, we first measured the amount of total inorganic carbon (TIC) by adding HCl into the sediment samples, with the HCl amount based on measured weight. By measuring the volume of CO2 produced, the amount of total inorganic carbon (TIC) was calculated. Then, the TOC was determined by subtracting the TIC from TC.

22

2  Research Area and Research Methods

When the climate was warm and humid, the forest generally grew better than the grasses, and the value of C/N was higher. In addition, hydrological activities were stronger when precipitation was higher, and more terrestrial soil material were therefore brought to the basin, resulting in a higher C/N value (O’Reilly et al. 2003). At the same time, the lake level change could have caused sediment disturbance, promoting the decomposition of organic matter (Kernan et al. 2010). When the climate was humid, the disturbance was low, generating high C/N values. In summary, C/N (TOC/TN) can be used as an indicator of climate, with a higher C/N indicating a warm and humid environment and a lower C/N indicating a dry and cold environment (Rosén and Hammarlund 2007; Williams et al. 2009). In addition, a higher TOC value alone indicates a warmer and more humid climate (Rosén 2005), particularly higher precipitation, in the semiarid region of China (Xiao et al. 2006). 2.3.1.5  Loss on Ignition (LOI) LOI was measured by the weight difference of approximately 1 g sediment sample before and after the sample was placed into a muffle furnace and baked at 980 °C for 30 min.

LOI% = ( W2 − W3 ) / ( W2 − W1 ) × 100%



(2.3)

W1: weight of empty crucible W2: weight of crucible with added sample W3: weight of crucible with added sample after heating in the muffle furnace The LOI reflected both the organic matter and CaCO3 content, which generally coincided with the fine grain size and TOC value. 2.3.1.6  Constant and Trace Elements The samples were first dried at 105 °C for 12 h to remove the capillary water, and boric acid was added to an aluminum bowl. Then, a small amount of the samples was added to the boric acid to fill the bowl. Finally, the aluminum bowl was placed on a tablet machine. K2O, Na2O, CaO, MgO, Fe2O3 and MnO were measured by a fluorescence spectrometer (ADVANT XP+) with the sample tablets. The accumulation of different elements in the soil differs under various environmental conditions, so the proportion of different chemical elements can reflect climate change. Generally, K2O, Na2O, CaO, and MgO are enriched in relatively dry soil, while P2O5, Fe2O3, TiO2, Al2O3, and MnO are enriched in soil under humid climate conditions (Liu et al. 2002b; Yin et al. 2012). The element ratio (ER) is used as an indicator in this paper.

2.3  Research Methods



ER = ( K 2 O + Na 2 O + CaO + MgO ) / ( Fe 2 O3 + MnO )

23



(2.4)

The greater the ratio, the more arid the climate is, and a small ratio indicates that the climate is wetter.

2.3.2  Establishment of a Regional Pollen Database 2.3.2.1  Sediment Data in the Forest-Steppe Ecotone in Northern China A total of nine original sites (Anguli Nuur, Hulun Nuur, Huangqihai, Liuzhouwan, Jiangjunpaozi, Xiaoniuchang, Ulan Nuur, Bai Nuur, and Haoluku) were cored and analyzed by our research group (Liu et al. 2001; Hao et al. 2014; Wang et al. 2012; Yin et al. 2015) following the methods detailed in Hao et al. (2014) and Yin et al. (2015). We also digitized published pollen data to better study the regional palynological composition patterns. This paper collected pollen data from 143 Holocene sedimentary sections in northern China (95°~130°E, 30°~50°N) and used Getdata Graph Digitizer 2.24 software (http://www.getdata-graph-digitizer.com) to digitize the total percentages of major tree genera (Pinus, Quercus, Betula) and arboreal pollen (AP). All radiocarbon dates were converted to calibrated years before the present (cal year BP) using the IntCal13 calibration data set implemented in Calib7.0.4 (Reimer et  al. 2013). This study uses the age model results of the original paper when possible and the piecewise linear fitting method to establish the age model if the original paper did not contain an age model. The precision, accuracy, and pollen sample dating methods of the 143 deposition profile data are not consistent. For different research contents, according to the research needs, the record number and position were illustrated, particularly in the following chapters. 2.3.2.2  Topsoil Pollen Assemblages in Northern China This study collected data on topsoil pollen assemblages in northern China, including Inner Mongolia and Hebei, Shanxi, and Gansu provinces. 2.3.2.3  Regional Climate Indexes of the Holocene Stalagmites have been used as an ideal material for past climate research due to their continuous deposits (continuous deposition for more than 100,000 years), high resolution (the oxygen and carbon isotopic resolution can reach even the seasonal or interannual scale), absolute dating accuracy (test error of 230Th up to 0.1%), and small alteration. In the analysis of the relationship between vegetation and climate,

24

2  Research Area and Research Methods

this paper chose carbon and oxygen isotope records of stalagmites as the main indicators of East Asian monsoon records, with records such as those from Dongge Cave (Dykoski et al. 2005; Wang et al. 2005), Sanbao Cave (Shao et al. 2006; Wang et al. 2008), and Heshang Cave (Hu et al. 2008). This evidence indicates that the summer monsoon activity gradually intensified in the early Holocene and then declined. The peak occurred at approximately 9.4~3.1 cal ka BP (Zhao et al. 2009). For research at regional and local scales, this paper also collected the physical and chemical properties of the sediment records in the study area with high precision, including the grain sizes of Daihai Lake and Hulun Nuur and sand dune dynamics of Mu Us and Otindag Sandy Land. These properties could be used to reflect local climate change, especially precipitation changes.

2.3.3  Phylogeography (Cytoplasmic DNA Analysis) 2.3.3.1  Sample Collection New foliar samples of tree species were collected from 8 to 20 individuals spaced ≥50 m apart at each population across the entire natural range. Each sample was deposited in a plastic bag with silica gel and stored in a refrigerator until DNA extraction. 2.3.3.2  DNA Extraction, Amplification, and Sequencing For each sample, total DNA was extracted from approximately 10 mg of silica gel-­ dried needles according to a hexadecetyltrimethylammonium bromide (CTAB) procedure (Doyle and Doyle 1987). The mtDNA or cpDNA fragments were amplified and sequenced using primers. Polymerase chain reaction (PCR) was performed in a 30  μL volume, containing 10–40  ng of DNA (ca. 2  μL) for each sample, 15  μL 2 × Taq MasterMix (CW0682; Beijing ComWin Biotech Co., Ltd., Beijing, China), 0.5 μL of each primer, and 12 μL of ddH2O. The PCR conditions for both mtDNA and cpDNA analyses followed the protocol provided by Chen et al. (2008). 2.3.3.3  Data Analysis Sequences were aligned using Vector NTI (Lu and Moriyama 2004). The relationships among mitotypes and chlorotypes were visualized using the median-joining network algorithm implemented in Network version 4.6.1.1 (Bandelt et al. 1999). Spatial genetic structure among the populations was analyzed for both mtDNA and cpDNA with (1) the Bayesian algorithm implemented in BAPS 6.0 (Corander et al. 2008) and (2) a simulated annealing approach using SAMOVA 2.0 (Dupanloup et al. 2002) with the number of groups set between one and eight. The SAMOVA

2.3  Research Methods

25

grouping was considered to be optimal when the differentiation among groups (FCT) reached a plateau. Analyses of molecular variance (AMOVA; Excoffier et al. 1992) were performed in Arlequin 3.5 (Excoffier and Lischer 2010) to assess the genetic differentiation within and among populations based on the groups inferred by BAPS analyses on mtDNA and cpDNA. Within-population gene diversity (HS), total gene diversity (HT), and population differentiation with unordered and ordered alleles (GST and NST, respectively) were estimated for mtDNA and cpDNA, using Permut (Pons and Petit 1996). NST > GST (after 1000 permutations) indicates the presence of phylogeographical structure with closely related haplotypes being found more often in a given geographical area than would be expected by chance (Pons and Petit 1996). A geographical map of the genetic differentiation among populations was obtained using the “interpolated genetic landscape shapes” procedure implemented in Alleles In Space (AIS) software (Miller 2005). The procedure first builds a connectivity network between all sampled locations in the data set based on Delaunay triangulations. Then, each network connection is assigned a genetic distance based on the average proportions of nucleotide differences between pairs of individuals from contiguous populations. Finally, genetic distances are interpolated (inverse distance weighted) to obtain relative genetic distances across the geographical landscape. Interpolation was conducted with a grid of 200 × 200 and a distance weighting parameter of 0.25. This analysis was conducted using mtDNA data to illustrate patterns of population differentiation that could not be easily visualized but was omitted for cpDNA because of the little genetic variation.

2.3.4  Present and LGM Species Distribution Modeling The past and current potential distributions of these four tree species were simulated using the species distribution model implemented in MaxEnt 3.3.3 (Phillips et al. 2006) for the present (1950–2000) and the LGM (22,000 cal year BP). MaxEnt is a machine learning-based ecological niche model that uses presence points and performs well even with few occurrence points (Elith et al. 2006). In addition to the presence records by our field studies, other records were obtained from the published data set (Mao and Wang 2011) available online (https://doi.org/10.5061/ dryad.8075) or on the Chinese Virtual Herbarium (http://www.cvh.org.cn/). We extracted the 19 bioclimatic WorldClim data layers at 2.5 arcmin resolution (Hijmans et  al. 2005; http://www.worldclim.org) and elevation data from the ASTER Global Digital Elevation Model (http://www.gdem.aster.ersdac.or.jp/ search.jsp) (Table  2.1). Highly correlated climatic variables (r  >  0.85) as well as those that contributed little to the predictive power (i.e., < 5% contribution) in trial runs of the SDMs were removed from subsequent analyses. About 80% of species distribution points were used as training data, and 20% of species distribution points were used as test data. The model was repeated ten times. The model of the species’ present distribution thus used three to five bioclimatic data layers. The best performing species distribution model was projected onto the CCSM, MPI, and MIROC

26

2  Research Area and Research Methods

Table 2.1  Twenty factors used in MaxEnt modeling bio1 bio2 bio3 bio4 bio5 bio6 bio7 bio8 bio9 bio10 bio11 bio12 bio13 bio14 bio15 bio16 bio17 bio18 bio19 dem_China

Environmental variables Mean annual temperature Mean diurnal range[mean of monthly (max temperature-min temperature)] Isothermality (bio2/bio7 × 100) SD of temperature seasonality Max temperature of warmest month Min temperature of coldest month Temperature annual range (bio5–bio6) Mean temperature of wettest quarter Mean temperature of driest quarter Mean temperature of warmest quarter Mean temperature of coldest quarter Annual precipitation Precipitation of wettest month Precipitation of driest month Coefficient of variation of precipitation seasonality Precipitation of wettest quarter Precipitation of driest quarter Precipitation of warmest quarter Precipitation of coldest quarter DEM

climate data layers (Collins et al. 2006) for the LGM (http://www.worldclim.org/ paleo-climate). The accuracy and credibility of the model predictions were assessed by calculating the area under the curve (AUC; Fawcett 2006).

References An, C., Feng, Z., Barton, L. 2006. Dry or humid? Mid-Holocene humidity changes in arid and semi-arid China. Quaternary Science Reviews, 25, 351–361. Bandelt, H.J., Forster, P., Röhl, A. 1999. Median-joining networks for inferring intraspecific phylogenies. Molecular Biology and Evolution, 16, 37–48. Collins, W.D., Bitz, C.M., Blackmon, M.L., Bonan, G.B., Bretherton, C.S., Carton, J.A., Chang, P., Doney, S.C., Hack, J.J., Henderson, T.B., Kiehl, J.T., Large, W.G., McKenna, D.S., Santer, B.D., Smith, R.D. 2006. The Community Climate System Model Version 3 (CCSM3). Journal of Climate, 19, 2122–2143. Corander, J., Sirén, J., Arjas, E. 2008. Bayesian spatial modeling ofgenetic population structure. Computational Statistics, 23, 111–129. De Chen, K., Abbott, R.J., Milne, R.I., Tian, X.M., Liu, J. 2008. Phylogeography of Pinus tabulaeformis Carr. (Pinaceae), a dominant species of coniferous forest in northern China. Molecular Ecology, 17, 4276–4288. Deckker, P., Corrège, T., Head, J.  1991. Late Pleistocene record of cyclic eolian activity from tropical Australia suggesting the Younger Dryas is not an unusual climatic event. Geology, 19, 602–605.

References

27

Doyle, J.J., Doyle, J.L. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin, 19, 11–15. Dupanloup, I., Schneider, S., Excoffier, L. 2002. A simulated annealing approach to define the genetic structure of populations. Molecular Ecology, 11, 2571–2581. Dykoski, C.A., Edwards, R.L., Cheng, H., Yuan, D., Cai, Y., Zhang, M., Lin, Y., Qing, J., An, Z., Revenaugh, J. 2005. A high-resolution, absolute-dated Holocene and deglacial Asian monsoon record from Dongge Cave, China. Earth and Planetary Science Letters, 233, 71–86. Elith, J., Graham, C. H., Anderson, R. P., et al. 2006. Novel methods improve prediction of species’ distributions from occurrence data. Ecography, 29, 129–151. Excoffier, L., Smouse, P.E., Quattro, J.M. 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics, 131, 479–491. Excoffier, L., Lischer, H.E.L. 2010. Arlequin suite ver 3.5: a new series of programs to perform population genetic analyses under Linux and Windows. Molecular Ecology Resources, 10, 564–567. Fawcett, T. 2006. An introduction to ROC analysis. Pattern Recognition Letters, 27, 861–874. Hao, Q., Liu, H., Yin, Y., Wang, H., Feng, M. 2014. Varied responses of forest at its distribution margin to Holocene monsoon development in northern China. Palaeogeography, Palaeoclimatology, Palaeoecology, 409, 239–248. Herzschuh, U. 2007. Reliability of pollen ratios for environmental reconstructions on the Tibetan Plateau. Journal of Biogeography, 34, 1265–1273. Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G., Jarvis, A. 2005. Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology, 25, 1965–1978. Hou, X. 1988. Natural geography of China: Plant geography (Volume II). Beijing: Science Press. Huang, X., Chen, F., Fan, Y., Yang, M. 2009. Dry late-glacial and early Holocene climate in arid central Asia indicated by lithological and palynological evidence from Bosten Lake, China. Quaternary International, 194, 19–27. Hu, C., Henderson, G.M., Huang, J., Xie, S., Sun, Y., Johnson, K.R. 2008. Quantification of Holocene Asian monsoon rainfall from spatially separated cave records. Earth and Planetary Science Letters, 266, 221–232. Kernan, M., Battarbee, R.W., and Moss, B. 2010. Climate change impacts on freshwater ecosystems, Wiley Online Library. Liang, E., D. Eckstein. 2006. Light rings in Chinese pine (Pinus tabulaeformis) in semiarid areas of north China and their palaeo-climatological potential. New Phytologist, 171, 783–791. Liu, H., Cui, H., Huang, Y. 2001. Detecting Holocene movements of the woodland–steppe ecotone in northern China using discriminant analysis. Journal of Quaternary Science, 16, 237–244. Liu, H., Cui, H., Yu, P., Huang, Y. 2002a. The origin of remnant forest stands of Pinus tabulaeformis in southeastern Inner Mongolia. Plant Ecology, 158, 139–151. Liu, H., Xu, L., Cui, H. 2002b. Holocene history of desertification along the woodland-steppe border in northern China. Quaternary Research, 57, 259–270. Liu, H., Yin, Y., Zhu, J., Zhao, F., Wang, H. 2010. How did the forest respond to Holocene climate drying at the forest-steppe ecotone in northern China? Quaternary International, 227, 46–52. Liu, Y., H. W. Linderholm, H. Song, Q. Cai, Q. Tian, J. Sun, D. Chen, E. Simelton, K. Seftigen, H.  Tian. 2009. Temperature variations recorded in Pinus tabulaeformis tree rings from the southern and northern slopes of the central Qinling Mountains, central China. Boreas, 38, 285–291. Lu, G., Moriyama, E.N. 2004. Vector NTI, a balanced all-in-one sequence analysis suite. Briefings in Bioinformatics, 5, 378–388. Maher, L.J., 1981. Statistics for microfossil concentration measurements employing samples spiked with marker grains. Review of Palaeobotany and Palynology, 32, 153–191. Mao, J.-F., Wang, X.-R. 2011. Distinct Niche Divergence Characterizes the Homoploid Hybrid Speciation of Pinus densata on the Tibetan Plateau. The American Naturalist, 177, 424–439.

28

2  Research Area and Research Methods

Miller, M.P. 2005. Alleles In Space (AIS): Computer Software for the Joint Analysis of Interindividual Spatial and Genetic Information. Journal of Heredity, 96, 722–724. Moore, P.D., Webb, J.A., Collison, M. 1991. Pollen analysis. Blackwell Scientific Publications, Oxford, UK. O’Reilly, C.M., Alin, S.R., Plisnier, P.D., Cohen, A.S., McKee, B.A. 2003. Climate change decreases aquatic ecosystem productivity of Lake Tanganyika, Africa. Nature, 424, 766–768. Phillips, S.J., Anderson, R.P., Schapire, R.E. 2006. Maximum entropy modeling of species geographic distributions. Ecological Modelling, 190, 231–259. Pons, O., Petit, R. 1996. Measuring and testing genetic differentiation with ordered versus unordered alleles. Genetics, 144, 1237–1245. Porter, S.C., An, Z. 1995. Correlation between climate events in the North Atlantic and China during the last glaciation. Nature, 375, 305–308. Qiang, M., Chen, F., Zhang, J., Zu, R., Jin, M., Zhou, A., Xiao, S. 2007. Grain size in sediments from Lake Sugan: a possible linkage to dust storm events at the northern margin of the Qinghai-­ Tibetan Plateau. Environmental Geology, 51, 1229–1238. Qin, D. 2005. Climate and environmental evolution of China: The evolution and prediction of climate and environment. Beijing: Science Press. Reimer, P.J., Bard, E., Bayliss, A., et al. 2013. IntCal13 and Marine13 radiocarbon age calibration curves 0–50,000 cal years BP. Radiocarbon, 55, 1869–1887. Rosén, P. 2005. Total organic carbon (TOC) of lake water during the Holocene inferred from lake sediments and near-infrared spectroscopy (NIRS) in eight lakes from northern Sweden. Biogeochemistry, 76, 503–516. Rosén, P., Hammarlund, D. 2007. Effects of climate, fire and vegetation development on Holocene changes in total organic carbon concentration in three boreal forest lakes in northern Sweden. Biogeosciences, 4, 975–984. Shao, X., Wang, Y., Cheng, H., Kong, X., Wu, J., Edwards, R.L. 2006. Long-term trend and abrupt events of the Holocene Asian monsoon inferred from a stalagmite δ18O record from Shennongjia in Central China. Chinese Science Bulletin, 51, 221–228. Shi, J., Liu, Y., Vaganov, E.A., Li, J., Cai, Q. 2008. Statistical and process-based modeling analyses of tree growth response to climate in semi-arid area of north central China: A case study of Pinus tabulaeformis. Journal of Geophysical Research: Biogeosciences, 113, 2005–2012. Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W., Burr, C., Hughen, K.A., Kromer, B., McCormac, G., Plicht, J.v.d., Spurk, M. 1998. INTCAL98 Radiocarbon Age Calibration, 24,000–0 cal BP. Radiocarbon. International Journal of Cosmogenic Isotope Research, 40, 1041–1084. Wang, Y., Cheng, H., Edwards, R., He, Y., Kong, X., An, Z., Wu, J., Kelly, M., Dykoski, C., Li, X. 2005. The Holocene Asian monsoon: links to solar changes and North Atlantic climate. Science, 308, 854. Wang, Y., Cheng, H., Edwards, R.L., Kong, X., Shao, X., Chen, S., Wu, J., Jiang, X., Wang, X., An, Z. 2008. Millennial-and orbital-scale changes in the East Asian monsoon over the past 224,000 years. Nature, 451, 1090–1093. Wang, H., Liu, H., Zhao, F., Yin, Y., Zhu, J., Snowball, I. 2012. Early- and mid-Holocene palaeoenvironments as revealed by mineral magnetic, geochemical and palynological data of sediments from Bai Nuur and Ulan Nuur, southeastern inner Mongolia Plateau, China. Quaternary International, 250, 100–118. Williams, J.W., Shuman, B., Bartlein, P.J. 2009. Rapid responses of the prairie-forest ecotone to early Holocene aridity in mid-continental North America. Global and Planetary Change, 66, 195–207. Wu, Y., Lücke, A., Zhangdong, J., Sumin, W., Schleser, G.H., Battarbee, R.W., Weilan, X. 2006. Holocene climate development on the central Tibetan Plateau: a sedimentary record from Cuoe Lake. Palaeogeography, Palaeoclimatology, Palaeoecology, 234, 328–340. Wu, Z. 1980. The vegetation of China. Beijing: Science Press. Xiao, J., Wu, J., Si, B., Liang, W., Nakamura, T., Liu, B., Inouchi, Y. 2006. Holocene climate changes in the monsoon/arid transition reflected by carbon concentration in Daihai Lake of Inner Mongolia. The Holocene, 16, 551–560.

References

29

Xu, H. 1990. Pinus tabuliformis Carrière. Beijing: China Forestry Publishing House. Yin, Y. 2012. Holocene forest evolution and its driving factors in semi-humid and semi-arid regions of China. Thesis for Ph.D. degree of Peking University. Yin, Y., Liu, H., Hao, Q. 2015. The role of fire in the late Holocene forest decline in semi-arid North China. The Holocene, 26, 93–101. Yin, Y., Liu, H., He, S., Zhao, F., Zhu, J., Wang, H., Liu, G., Wu, X. 2011. Patterns of local and regional grain size distribution and their application to Holocene climate reconstruction in semi-arid Inner Mongolia, China. Palaeogeography, Palaeoclimatology, Palaeoecology, 307, 168–176. Yin, Y., Liu, H., Liu, G., Hao, Q., Wang, H. 2012. Vegetation responses to mid-Holocene extreme drought events and subsequent long-term drought on the southeastern Inner Mongolian Plateau, China. Agricultural and Forest Meteorology, 178–179, 3–9. Zhai, Q.M., Guo, Z.Y., Li, Y.L., Li, R.Q., 2006. Annually laminated lake sediments and environmental changes in Bashang Plateau, North China. Palaeogeography, Palaeoclimatology, Palaeoecology, 241, 95–102. Zhao, Y., Yu, Z. 2012. Vegetation response to Holocene climate change in East Asian monsoon-­ margin region. Earth-Science Reviews, 113, 1–10. Zhao, Y., Yu, Z., Chen, F. 2009. Spatial and temporal patterns of Holocene vegetation and climate changes in arid and semi-arid China. Quaternary International, 194, 6–18. Zhao, Y., Yu, Z., Chen, F., Ito, E., Zhao, C. 2007. Holocene vegetation and climate history at Hurleg Lake in the Qaidam Basin, northwest China. Review of Palaeobotany and Palynology, 145, 275–288. Zhou, T. 1992. Holocene environmental evolution and prediction in the transition agriculture-­ animal husbandry zone of north China. Beijing: Geological Publishing House.

Chapter 3

Glacial Refugia and the Postglacial Migration of Dominant Tree Species in Northern China

In this study, we used an integrated approach to investigate the most likely location of glacial refugia and the postglacial migration patterns of four dominant tree species (Pinus tabulaeformis, Quercus mongolica, Betula platyphylla, and Betula dahurica). Specifically, we surveyed range-wide cytoplasmic DNA variations in order to refine the phylogeographical history of P. tabulaeformis. An analysis of published and original pollen records was used to assess the glacial range and map the chronology of postglacial migrations of Pinus, Quercus, and Betula. Ecological niche modeling was applied to identify potential distribution of these species during the LGM.  Through the above three aspects of evidence, this chapter would test the hypothesis 1: During the LGM, Chinese pine has refugia in northern China. Northern mountains and the sandy land may be the refugia. For other three species, the northern China could also supply LGM refugia. It is further speculated that the modern northern forest populations are not necessarily the result of long-distance migration from low latitudes but also the result of local diffusion of populations in high-­ latitude refugia. Combining these three lines of evidence provided new insights into the Quaternary biogeography of this widespread species in China.

3.1  G  lacial Refugia and the Postglacial Migration of Pinus tabulaeformis 3.1.1  Distribution Change Based on Pollen Data Analysis We initially obtained pollen records from 135 core sites located north of the Yangtze River, northern China. We used the following criteria to select 42 published pollen records for inclusion in this study (Fig.  3.1): (1) geographical coordinates were within a polygon delimited by longitude 95°–130°E and latitude 30°–45°N, which © Springer Nature Singapore Pte Ltd. 2018 Q. Hao, The LGM Distribution of Dominant Tree Genera in Northern China’s Forest-steppe Ecotone and Their Postglacial Migration, Springer Theses, https://doi.org/10.1007/978-981-13-2883-1_3

31

32

3  Glacial Refugia and the Postglacial Migration of Dominant Tree Species…

Fig. 3.1  Location of the study area at regional and local scales. (a) Map of China showing location of study region (rectangle is detailed in b), pollen record sites, and current natural range of Pinus tabulaeformis (green). (b) The 20 new sampling sites for cytoplasmic DNA analyses are shown as dark green dots and 30 sites from Chen et al. (2008) are shown as brown dots. The main geographical features discussed in the manuscript (Loess Plateau, Qinling Mountains, Qilian Mountains, Taihang Mountains, Otindag Sandy Land) are shown on the map. Field photographs of P. tabulaeformis from Otindag Sandy Land and the mountainous areas are shown on the right. (Reproduced from with permission Hao et al. (2018). Copyright (2017) John Wiley and Sons)

covers the entire natural range of P. tabulaeformis; (2) mean sampling resolution was 30%. Between 16,000 and 12,000 cal year BP, the Pinus pollen percentage at Huangjiapu (located close to the southeastern edge of the Otindag Sandy Land) was higher than 30%. The range of the species expanded in central Inner Mongolia and east of Gansu province, at the northern edge of its modern distribution. Between 12,000 and 8000 cal year BP, Pinus pollen reached >10% in 14 records of the mountainous areas and the Otindag Sandy Land. Pinus was widely distributed during this period throughout northern China, where the taxon constituted 20–30% of the pollen assemblages even in today’s semiarid areas. In the 42 sediment pollen records in the study, this research thought that the Pinus indicates the changes of Chinese pine based on their distribution (Fig. 3.3). Although the northern of Pinus armandii and southwest of Pinus tabulaeformis overlap partly (Fig. 3.3), our study area is mainly dominated by Pinus tabulaeformis. Therefore, this study suggests that pine pollen should be derived mainly from Pinus tabulaeformis, and its uncertainty will be discussed in the Chap. 6. The previous studies tried to make distinction among different species in the genus of Pinus (Jiang 1984), but it is still difficult to identify different species pollen in the sediments, and previous research has not been documented the morphological differences between species.

3.1  Glacial Refugia and the Postglacial Migration of Pinus tabulaeformis

35

3.1.2  Genetic Diversity Within and Among P. tabulaeformis Populations In the Pinaceae, mitochondrial DNA (mtDNA) is the maternal inheritance carried by seed, while chloroplast DNA (cpDNA) is paternal inheritance with pollen and seed as carriers (Neale and Sederoff 1989; Wagner 1992). The difference with angiosperm is mainly related to the process of fertilization of gymnosperm. Neale et al. (1986) firstly use the RFLP technology in the Pseudotsuga menziesii observing paternal inheritance of chloroplast DNA in hybrid progenies. In the 1980s, studies on several gymnosperm cytoplasms were observed under the electron microscope in the process of fertilization, zygote, and embryo developmental stage, putting forward the concept of “neocytoplasm,” new cytoplasm produced in the oosperm. The organelles in this cytoplasm are not only from the mitochondria of oocyte but also from the mitochondria and plastids in the cytoplasm of sperm. In the oosperm, only the neocytoplasm forms the cytoplasm of the embryo, and the cytoplasm in the oocyte gradually degenerates (Li and Gao 2008). The long-distance spread of pollen (such as Pinaceae) will lead to higher intraspecific gene flow in chloroplast DNA than mitochondrial DNA (i.e., less obvious genetic structure) (Burban and Petit 2003; reviewed in Du et al. 2009). Studies have shown that the method with chloroplast markers can be used to study the transcontinental range changes of North American coniferous species in the distribution of the Quaternary (Anderson et al. 2006; de Lafontaine et al. 2010; Gérardi et al. 2010; Cinget et al. 2015; Warren et al. 2016), but the long-distance pollen dispersal may erode gene signal, resulting in less pronounced genetic structure in smaller geographic range, due to the genetic isolation (Liepelt et al. 2002). In this study, we considered the genetic structure of both mitochondria and chloroplasts. In this study, 8~20 needles were picked from 20 natural populations of Pinus tabulaeformis in China (Table 3.1 and Fig. 3.1b). Each leaf sample is placed in a plastic bag containing silica gel and stored in a laboratory refrigerator. Two hundred fifty-three samples were used for chloroplast DNA (cpDNA) and mitochondrial DNA (mtDNA) analysis. The total DNA of each sample was extracted with silica gel dried needles (about 10  mg) by CTAB method (Doyle and Doyle 1987; Chen et  al. 2008). Using the primers in Table  3.2 (Chen et  al. 2008), two mitochondrial DNA fragments (nad5 intron 1 and nad4/3–4) and two chloroplast DNA fragments (rpl16和trnS-trnG) were amplified and sequenced. Polymerase chain reaction (PCR) was conducted in 30  μL, which contains 10~40 ng plant DNA (about 2 μL), 15 μL Taq MasterMix (CW0682; Beijing connwii century Biotechnology Co. Ltd.), 0.5  μL of each primer, and 12  μL double distilled water (Chen et al. 2008). The amplification process is as follows: mtDNA (nad5 intron 1and nad4/3–4), 94 °C 6 min—94 °C 45 s, 52 °C 45 s, 72 °C 2 min, 36 cycles—72 °C 7 min CpDNA (rpl16 and trnS-trnG). 94 °C 5 min—94 °C 50 s, 56 °C 50 s, 72 °C l min 10 s, 36 cycles—72 °C 8 min.

117.72 116.96 108.99 108.55

43.63 42.90 35.96 33.81

1097 1279 1015 1444

Altitude (m) 1254 1725 2552 1358 1467 1530 1369 1331 1397 1702 1438 2027 1067 1095 1412 1190 16 10 9 15 253

N 8 19 14 11 15 15 17 11 12 8 8 19 9 14 13 10 0 0 0 10 24

A 0 0 14 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 90

B 8 0 0 0 0 15 17 0 12 8 8 19 0 0 0 0 0 0 0 1 1

C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 13

D 0 0 0 0 0 0 0 0 0 0 0 0 0 0 13 0 16 10 6 0 120

E 0 19 0 11 15 0 0 11 0 0 0 0 8 14 0 10 0 0 3 1 5

F 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0

Mitotype (nad5, nad 4/3–4)

Thirteen populations are located in mountains (M) and seven are located in sandy land (SL)

Population Aguimiao (M) Baimatan (M) Changlingshan (M) Ganqinars (SL) Guandishan (M) Heshun (M) Jiangjiadian (SL) Jiukesong (SL) Lamagou (M) Qiliyu (M) Sandaohekou (SL) Suyukou (M) Tongchuan (M) Wazijie (M) Wutaishan (M) Xiangshuidianzhan (SL) XWNG Xiaowunigou (SL) YHT Yuanhetou (SL) ZL Zhiluo (M) ZQS Zhuqueshan (M) Total

ID AGM BMT CLS GNS GDS HS JJD JKS LMG QLY SDHK SYK TC WZJ WTS XSDZ

Longitude (°E) 110.70 110.02 103.70 116.34 111.73 113.36 117.58 116.61 111.29 112.00 116.99 105.92 108.77 110.01 113.76 117.04

Latitude (°N) 39.48 35.67 37.44 43.24 37.51 37.44 42.51 43.07 40.80 36.63 42.41 38.74 35.16 35.95 38.83 43.00 0.00 0.00 0.50 0.54

Gene diversity 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.22 0.00 0.00 0.00 12 6 5 13 204

4 4 2 1 35

0 0 0 0 7

0 0 0 0 2

0 0 2 1 4

0 0 0 0 4

0.40 0.53 0.67 0.26

Chlorotype (rpl16, trnS–trnG) Gene C1 C2 C5 C6 C7 C8 diversity 6 1 1 0 0 0 0.46 15 1 0 0 2 1 0.38 10 1 0 1 1 1 0.51 9 2 0 0 0 0 0.33 14 0 0 0 1 0 0.13 14 0 0 1 0 0 0.13 16 1 0 0 0 0 0.12 11 0 0 0 0 0 0.00 11 0 0 0 0 1 0.17 6 1 0 0 0 1 0.46 7 1 0 0 0 0 0.25 17 2 0 0 0 0 0.20 8 1 0 0 0 0 0.22 10 4 0 0 0 0 0.44 6 7 0 0 0 0 0.54 8 2 0 0 0 0 0.36

Table 3.1  Geographic and genetic characteristics of 20 P. tabulaeformis populations in our survey for mitochondria DNA and chloroplast DNA sequences variation

36 3  Glacial Refugia and the Postglacial Migration of Dominant Tree Species…

Genome mtDNA mtDNA cpDNA cpDNA

Name nad4/3–4 nad5 intron 1 rpl16 trnS-trnG

Forward primer (5′–3′) GTATTCCCCTTGGCTATATA TAGTCGGTGGAACCGGTGAA GCTATGCTTAGTGTGTGACTCGTTG GCCGCTTTAGTCCACTCAGC

Reverse primer (5′–3′) TGTTTACTACGATTCAGGGT TGAGGATGGACCAAGCTACT CCCTTCATTCTTCCTCTATGTTG GAACGAATCACACTTTTACCAC

References Dumolin-Lapegue et al. (1997) Jaramillo-Correa et al. (2003) Small et al. (1998) Demesure et al. (1995)

Table 3.2  Primers used in our study for mitochondria DNA and chloroplast DNA sequences extraction, amplification, and sequencing

3.1  Glacial Refugia and the Postglacial Migration of Pinus tabulaeformis 37

38

3  Glacial Refugia and the Postglacial Migration of Dominant Tree Species…

For data analysis, see Chap. 2. In order to compare with the previous studies, this paper unified the previous 30 populations with 20 populations collected by us (n = 50). The populations around sandy land on the northern edge of the distribution area were considered, and the population genetic diversity (HT) and population genetic diversity (HS) of Pinus tabulaeformis populations in seven sandy lands were calculated. Three mtDNA variants were found in nad4/3–4 and two in nad5 intron 1 (see Table  3.3), yielding a total of six mitotypes (A–F) across the 50 populations (Fig. 3.4a), all of which were previously reported by Chen et al. (2008). Mitotypes A and F had a relatively close genetic relationship, while the other four mitotypes formed another group in the network (Fig. 3.4a). Mitotypes B was the most common and was either fixed or dominant in 21 populations throughout the range. Mitotype E had a disjunct distribution. It was fixed or dominant in six populations from the mountains south of the Loess Plateau and completely fixed in five of the northernmost populations in the Sandy Land. Nine populations, all from the southwestern part of the range, were dominated by mitotype A. Mitotype F was common in southernmost and easternmost populations, whereas mitotype D was found in three populations located in the core of the range and was fixed in two of them. Mitotype C was rare and found at low frequency in the southern half of the range. Six variants were found in both cpDNA fragments (rpl16 and trnS-trnG). These were combined into a total of eight chlorotypes (C1–C8, Table 3.3). Chlorotypes C1–C5 were previously reported (Chen et al. 2008) and the other three were newly discovered in the present study. Chlorotype C1 occurred at the highest frequency and dominated all but two populations. Network analysis showed that all remaining chlorotypes had a close relationship with C1 except C4 that was closer to C5 (Fig. 3.4b). Chlorotype C2 occurred in 31 populations at lower frequency. All other chlorotypes were mainly restricted to the southern part of the range, except for chlorotype C3 found only in one of the northernmost populations from the Otindag Sandy Land. The Bayesian analysis of population structure revealed eight distinct groups of populations based on mtDNA variation (Fig. 3.5a). Each BAPS group had a distinct geographical distribution and was dominated by one or two specific mitotypes (Fig.  3.4a). Groups 1 and 5 were located in south central part of the range (i.e., Qinling Mountains) along with Group 2, which also extended eastward. Group 3 lumped together populations from the Otindag Sandy Land along with others from the Qinling Mountains, south of the Loess Plateau. Group 7 included the westernmost populations (i.e., Qilian Mountains) while the small Groups 4 and 6 included one and two populations from the west and central parts of the range (i.e., Qilian and Taihang Mountains), respectively. Finally, Group 8 was the most common and widely distributed, encompassing the core of the range up to the northernmost stands in the Otindag Sandy Land. The genetic structure of the cpDNA data provided limited information since all populations were clustered into one group except two singleton populations #27 and #30 (Fig. 3.5b). Based on SAMOVA, FCT estimates reached a plateau when the 50 sampled populations were clustered into five groups based on mtDNA data (Fig.  3.6). The

Nucleotide variable positions rpl16 Chlorotype 177 183~186 218 311 380 C1 A ---A A A C2 A ---A G A C3 A ---A A A C4 A GATA A A G C5 A ---A A A C6 C ---A A A C7 A ---T A A C8 A ---A A A Six mitotypes and eight chlorotypes of Pinus tabulaeformis were detected a TTCTCTATCTATTTAGGG

Mitotype A B C D E F

Nucleotide variable positions nad4/3–4 884~911 GGTGGGGGGGCTTATG-----------G GCCCCCCCAAATTAAGTCAAAAAAAGGG GCCCCCCCAAATTAAGTCAAAAAAAGGG GCCCCCCCAAATTAAGTAAAAAAAAGGG GCCCCCCCAAATTAAGTAAAAAAAAGGG GGTGGGGGGGCTTATG-----------G

610 A A C A A A A A

trnS-trnG 115 200~218 a C a C a C T – C – a C a C a C

204 C C C C C C C T

235 G G G – G G G G

247 A A A C A A A A

nad5 intron 1 212~239 GCCCCCCCAAATTAAGTCAAAAAAAGGG GGTGGGGGGGCTTATG-----------G GCCCCCCCAAATTAAGTCAAAAAAAGGG GCCCCCCCAAATTAAGTCAAAAAAAGGG GGTGGGGGGGCTTATG-----------G GGTGGGGGGGCTTATG-----------G

365 A A A C A A A A

Table 3.3  Variation positions of aligned sequences of two mtDNA fragments (nad4/3–4 and nad5) and two cpDNA fragments (rpl16 and trnS-trnG)

3.1  Glacial Refugia and the Postglacial Migration of Pinus tabulaeformis 39

40

3  Glacial Refugia and the Postglacial Migration of Dominant Tree Species…

Fig. 3.4  Frequencies of six mitotypes (a) and eight chlorotypes (b) in each population. Circle size proportional to sample size. Insets display the two haplotype networks with circle size proportional to the haplotype frequency. (Reproduced from with permission Hao et al. (2018). Copyright (2017) John Wiley and Sons)

3.1  Glacial Refugia and the Postglacial Migration of Pinus tabulaeformis

41

Fig. 3.5  Geographic structure of 50 populations based on BAPS and SAMOVA. (a) Circles of different colors indicate assignment of each population to one of the eight mtDNA BAPS groups and the five dashed lines represent SAMOVA clustering according to mitochondrial DNA data. (b) Circles of different colors indicate assignment of each population to one of the three cpDNA BAPS groups and the dashed lines represent SAMOVA clustering according to chloroplastic DNA data. (Reproduced from with permission Hao et al. (2018). Copyright (2017) John Wiley and Sons)

42

3  Glacial Refugia and the Postglacial Migration of Dominant Tree Species…

Fig. 3.6  The change of FCT according to the different grouping (k) values (mtDNA)

g­ rouping pattern was similar to the BAPS results reported above (Fig. 3.5a). The only differences were that BAPS mtDNA Groups 1, 3, and 8, all located in the north-­central portion of the species range, were lumped into one group by SAMOVA and that BAPS mtDNA Groups 4 and 5 were clustered into one SAMOVA group. SAMOVA indicated no genetic structuring for the cpDNA data apart from a singleton population (#30: Taishan) also identified by BAPS (Fig. 3.5b). AMOVA also corroborated the spatial patterns in the genetic structure defined by the Bayesian algorithm. A significant portion of mtDNA variance (86.01%) was contributed by differences among the eight BAPS groups, with only 0.62% and 13.37% accounted for by variation among populations within groups and within populations, respectively (Table 3.4). Similarly, 88.81% of cpDNA variation was found among BAPS groups, leaving 0.14% and 11.05% of variation among populations within groups and within populations, respectively (Table 3.4). Both GST and NST were high for mtDNA (both 0.82) but low for cpDNA (0.12 and 0.35, respectively) (Table 3.5). NST ≈ GST for mtDNA indicated the absence of significant phylogeographical structure. Indeed, the largest, north-central group was dominated by two mitotypes (B and E), and their distribution had no clear geographical pattern (Fig. 3.5a). NST > GST for cpDNA could reflect the higher haplotype diversity in the southern half as well as a slight shift in chlorotype C2 frequency toward higher values in the northeastern part of the range (Fig. 3.5b). The interpolated genetic landscape analysis illustrates that genetic (mtDNA) differentiation between adjacent populations was highest in the south central part of the range within the Loess Plateau and surrounding mountains (i.e., between 104 and 115°E and south of 38°N) but decreased rapidly outside of these geographic features (Fig. 3.7).

3.1  Glacial Refugia and the Postglacial Migration of Pinus tabulaeformis

43

Table 3.4 Hierarchical analysis of molecular variance (AMOVA) based on mitotype and chlorotype frequencies for 50 populations of Pinus tabulaeformis grouped according to the BAPS analysis based on mitochondrial DNA and chloroplastic DNA data, respectively Source variation mtDNA Among mtDNA BAPS groups Among populations within groups Within populations Total cpDNA Among cpDNA BAPS groups Among populations within groups Within populations Total

d.f.

SS

VC

Variation (%)

Fixation index

7 42 494 543

3922.47 92.57 723.66 4738.69

9.43 0.07 1.46 10.96

86.01 0.62 13.37

FCT = 0.860* FSC = 0.045* FST = 0.866*

2 47 494 543

141.42 26.68 245.58 413.68

4.00 0.01 0.50 4.50

88.81 0.14 11.05

FCT = 0.888* FSC = 0.013* FST = 0.890*

d.f. degrees of freedom, SS sum of squares, VC variance component, FCT differentiation among BAPS groups, FSC differentiation among populations within groups, FST differentiation within populations *P < 0.0001 Table 3.5  Estimates of average gene diversity within populations (HS), total gene diversity (HT), population differentiation with unordered (GST) and ordered alleles (NST) (mean ± SE in parentheses) for mitochondrial DNA and chloroplastic DNA Regions mtDNA Total distribution (50 populations) cpDNA Total distribution (50 populations)

HS

HT

GST

NST

0.14 (0.029)

0.74 (0.032)

0.82 (0.039)

0.82 (0.040)

0.25 (0.026)

0.28 (0.035)

0.12 (0.060)

0.35 (0.161)*

*P < 0.0001

3.1.3  S  imulated Present and LGM Distribution of  P. tabulaeformis The past and current potential distributions of P. tabulaeformis were simulated using the species distribution model implemented in MaxEnt 3.3.3 (Phillips et al. 2006) for the present (1950–2000) and the LGM (22,000 cal year BP). In addition to the 50 presence records in this and previous studies, 149 records were obtained from a published data set (Mao and Wang 2011) available online (https://doi. org/10.5061/dryad.8075). We extracted the 19 bioclimatic WorldClim data layers at 2.5 arcmin resolution (Hijmans et al. 2005; http://www.worldclim.org) and elevation data from ASTER Global Digital Elevation Model (http://www.gdem.aster.ersdac.or.jp/search.jsp) for the 199 presence records available. Highly correlated climatic variables (r > 0.85) as well as those that contributed little to the predictive power (i.e., < 5% contribution) in trial runs of the SDMs were removed from subsequent analyses. The model of the species’ present distribution thus used five

44

3  Glacial Refugia and the Postglacial Migration of Dominant Tree Species…

Fig. 3.7  Interpolated among-population relative genetic distances using mtDNA data across Pinus tabulaeformis range, conducted with a grid of 200 × 200 m. Higher (red) and lower (blue) relative genetic distance indicates stronger and lower differentiation among populations, respectively. Circles indicate the 50 populations and different colors represent their assignments to mtDNA BAPS groups, as indicated in Fig.  3.5a. (Reproduced from with permission Hao et  al. (2018). Copyright (2017) John Wiley and Sons)

bioclimatic data layers (BIO1 = mean annual temperature, BIO4 = temperature seasonality, BIO13 = precipitation of wettest month, BIO18 = precipitation of warmest quarter, and BIO19  =  precipitation of coldest quarter). The MaxEnt model was tuned with the R package ENMEval (Muscarella et  al. 2014). After partitioning occurrence data using the “checkerboard2” method, we built models with regularization multiplier values ranging from 0.5 to 4.0 (increments of 0.5) and with six different combinations of “feature classes” (L, LQ, H, LQH, LQHP, LQHPT; where L, linear; Q, quadratic; H, hinge; P, product; and T, threshold). Model selection was based on the Akaike information criterion corrected for small samples sizes (AICc; Warren and Seifert 2011) and by estimating the area under the “receiver operating characteristic” curve (AUC; Fawcett 2006). The best performing species distribution model was projected onto the CCSM and MIROC climate data layers (Collins et al. 2006) for the LGM (http://www.worldclim.org/paleo-climate). The maximum training sensitivity plus specificity (MTSS) logistic threshold was employed to establish areas where the species is likely to be present or absent (Liu et al. 2005, 2013). According to AICc, the best performing MaxEnt model was obtained with feature class combinations LQHPT with 2.0 as the regularization multiplier. The resulting AUC score for the species distribution modeling was high (0.97), indicating good predictive model performance (i.e., 1 is the maximum prediction and 0.5 sug-

3.1  Glacial Refugia and the Postglacial Migration of Pinus tabulaeformis

45

gests a random prediction). The predicted current distribution with logistic value above the MTSS threshold (0.065) was generally similar to, but slightly larger than, the observed distribution (Fig. 3.8a), along the 400 mm isohyet in northern China (Fig. 3.1). Except for local expansion/extirpation, the simulated LGM distribution (Fig. 3.8b, c) almost entirely overlapped with predicted current range (Fig. 3.8a). Indeed, during the LGM, the simulated distribution of P. tabulaeformis extended from 32 to 42°N and 102 to 124°E, centered on the Loess Plateau and some coastal region of northern China (Fig. 3.8b, c). There was no evidence of suitable climatic conditions south of 30°N, meaning that the entire glacial range likely persisted north of the Yangtze River. While it is unlikely that P. tabulaeformis could have persisted locally within the semiarid region (especially central Inner Mongolia) because the LGM climate was unsuitable there (distribution logistic value 0.75) in the area centered on the ZQS/Xipo populations and extending to all the mountainous areas surrounding the Loess Plateau (104 to 115°E and 32 to 38°N; compare Figs. 3.7 and 3.8b). This hindcasted distribution almost completely overlaps with the modern range, in sharp contrast with earlier phylogeographic inferences proposing low-­ elevation macrorefugia located within and beyond the southern part of the species’ current range (Chen et al. 2008). Indeed, the three lines of evidence reported here provide strong evidence that P. tabulaeformis likely existed in multiple local microrefugia during the LGM, all of which were located north of the Yangtze River, in the mountains bordering the southern and western margins of the Loess Plateau (i.e., Qilian, Ziwuling, and Qinling Mountains). Furthermore, mitotype D was endemic to the central part of the range (Taihang Mountains, bordering the eastern margin of the Loess Plateau) where it was fixed in two populations (forming mtDNA Group 6 in Fig. 3.5a). Although it is possible that this lineage persisted further south during the LGM, migrated northward, and disappeared from the south during the Holocene, this scenario seems unlikely because two extensive mtDNA groups (2 and 8) are found south of this highly spatially restricted lineage. Assuming that this haplotype persisted through the LGM in an isolated glacial mountain microrefugium, our data indicate that this glacial lineage did not spread during the postglacial period. Although the pattern is less clear for cpDNA, likely due to the homogenizing effect of pollen dispersal (Liepelt et  al. 2002; Burban and Petit 2003; Du et al. 2009), one of these populations (Wutaishan, WTS) is among the few that were not dominated by chlorotype C1. Besides, the paleodistribution model indicated a probability of presence >0.5, that is, well above the MTSS threshold (Fig. 3.8b), suggesting that this region was suitable for P. tabulaeformis during the LGM. No pollen record was found close to WTS during the LGM (Cao et al. 2015; Fig. 3.2). The earliest pollen record in the vicinity of this site dated back to 12,000 cal year BP with Pinus pollen already reaching ca. 15%, perhaps indicating an early presence of pine trees in this area (Fig. 3.2). In addition to the mountainous areas discussed above, pollen evidence suggests that Chinese pine forests might have persisted in the Otindag Sandy Land throughout the LGM.  Indeed, Pinus reached 38% in the Sanyi pollen record near the Otindag Sandy Land by 16,000 cal year BP and values >10% were maintained from 16,000 to 12,000  cal  year BP (Fig.  3.2). The Pinus pollen percentages of Dalai

48

3  Glacial Refugia and the Postglacial Migration of Dominant Tree Species…

Nuur, also located near the Otindag Sandy Land, were high during the LGM, ­reaching >35% from 22,000 to 12,000 cal year BP (Li et al. 1990; Cao et al. 2015). However, this pollen record was excluded from our analysis (i.e., this record does not appear in Fig. 3.2) because the sampling resolution was very low at this site (ca. 2000 year/sample). The percentages of Pinus pollen generally reached values >10% after ca. 9000 cal year BP in semiarid regions located at the northern edge of the current distribution of P. tabulaeformis. Yet, values >10% were reported at 11,000  cal  year BP at the northern semiarid pollen site Anguli Nuur (Liu et  al. 2010), indicating the presence of Pinus at least 2000 years earlier than at other semiarid sites. Other pollen sites located close to Anguli Nuur maintained high Pinus percentages (> 10%) after 9000 cal year BP, while Pinus pollen abundance did not increase in areas south of Anguli Nuur. This pattern suggests that Pinus in the vicinity of Anguli Nuur did not originate from long-distance postglacial migration from the south. Dalai Nuur, Sanyi, and Anguli Nuur are all located somewhat close to the Otindag Sandy Land. Although the sampling resolution was not high for Sanyi (660 years/sample; five radiocarbon and OSL dates for the entire record) or Dali Nuur (> 2000 years/sample), the sampling resolution was high and the chronology was robust for Anguli Nuur (Liu et al. 2010). According to the species distribution model, the LGM macroclimate was suitable for the persistence of P. tabulaeformis in the southern part of Otindag Sandy Land. Note that the present species distribution model indicated a low probability of current Chinese pine presence in the Otindag Sandy Land area (Fig. 3.8a), yet P. tabulaeformis is found at low density throughout the area today. Dry climate might have restricted pine forest more than low temperature during the LGM in this region where the modern landscape is dominated by semiarid steppe (Liu et al. 2004; Li et al. 2007). The current presence of P. tabulaeformis in the Otindag Sandy Land likely results from high groundwater levels of this region, which could supply enough groundwater for P. tabulaeformis. It is possible that groundwater was also sufficient for maintaining P. tabulaeformis in the region during the LGM. Hence, based on pollen records and modeling result, the hypothesis of a local spread from refugial populations in the Otindag Sandy Land and adjacent areas cannot be rejected. The seven populations sampled in the Otindag Sandy Land did not stand out as a region of high genetic diversity or endemism for mtDNA (Fig. 3.4a). However, the disjunct distribution of mitotype E (mtDNA, Group 3) could provide support for glacial persistence in the Sandy Land. Such a division in the distribution of a genetic variant (~1000 km apart) would be expected if the species range became disconnected during the LGM when the species persisted in two isolated refugia. While one was likely in the Qinling Mountains as previously discussed, the other would have to be further north, within—or at least close to—the Sandy Land. Keller et al. (2010) reported similar levels of low genetic diversity in remote Alaskan populations of balsam poplar, although fossil evidence indicated that the area was indeed a glacial refugium for this species (see also Breen et al. 2012). An endemic chlorotype (C3) was endemic to a population at the northernmost limit of the species distribution (Fig. 3.4b) providing genetic support for a northern refugium.

3.2  Glacial Refugia and the Postglacial Migration of Quercus mongolica

49

Some discrepancy between inferences from fossil records, phylogeography, and ecological niche models is expected at a local scale (Hugall et al. 2002) because these lines of evidence vary in their sensitivity and specificity (Gavin et al. 2014). The coarse modeling approach implemented in MaxEnt provides broad-scale boundaries of macrorefugia (Temunović et al. 2013). Even if the regional macroclimate was unsuitable for regional survival of P. tabulaeformis, specific microclimates in complex topographies might have provided suitable habitats for its local persistence (Dobrowski 2011; Ashcroft et al. 2012; de Lafontaine et al. 2014; Birks 2015). Thus, the question regarding glacial persistence in Otindag Sandy Land remains open. The mixed evidence reported here highlights the need for further genetic and palaeoecological studies focusing on this area characterized by a complex geography. Because the species was distributed in multiple glacial refugia across a large area north of the Yangtze River, encompassing several mountains of northern China and possibly the Otindag Sandy Land, its postglacial colonization occurred via local spread in multiple directions. This scenario stands in sharp contrast with a unidirectional long-distance south-to-north postglacial migration scenario suggested in previous studies (Yu et al. 1998; Ni et al. 2010), including an earlier phylogeographical investigation on the same species (Chen et  al. 2008). Our results contribute to a growing body of evidence suggesting multiple glacial refugia in northern China likely played a major role in the postglacial development of temperate forests in the region.

3.2  G  lacial Refugia and the Postglacial Migration of Quercus mongolica 3.2.1  Distribution Change Based on Pollen Data Analysis Quercus pollen is locally distributed. Even in deciduous oak forests in northern China, the pollen percentage of pine (9.6–46.3%) may exceed that of Quercus (1.7– 31.8%) (Xu et al. 2005). The pollen content of Quercus is generally lower than 2% in meadow in northern China (Liu et al. 2002). Xu et al. (2007) suggested that 2.5% could be used as a threshold for Quercus existence. Ren and Beug (2002) used 3% and 30% as thresholds for the presence and dominance of Quercus, while Cao et al. (2015) used > 6% as thresholds for the presence. Among modern European samples, Lisitsyna et al. (2011) summarized previous studies and obtained thresholds of 1.5–10% for modern pollen and 0–2% for sediment pollen. Milecka et al. (2004) and Huntley and Birks (1983) considered that 2% could be used as a threshold for the existence of Quercus, while Lisitsyna et al. (2011) used 1.5% as a threshold. In this study, 3% is used to as the threshold of Quercus presence. During the LGM, there were relatively few pollen records (n = 4), and the pollen percentage of Quercus was more than 3% in Huanghua and Suancigou. During the

50

3  Glacial Refugia and the Postglacial Migration of Dominant Tree Species…

18,000–15,000 cal year BP, more than 3% of Quercus pollen was found in the Yan Mountains in northern Hebei Province, and higher percentage was found in the Suancigou region of Gansu Province. Since 15,000 cal year BP, pollen records gradually increased and the distribution range gradually expanded. During the early Holocene, the pollen percentage of Quercus was low. Quercus began to appear in eastern Liaoning Province. From 9000 to 6000 years ago, the distribution range and coverage reached the maximum and widely distributed in northern and northeastern China. Generally, the distribution range of Quercus increased firstly and then decreased.

3.2.2  S  imulated Present and LGM Distribution of  Q. mongolica According to the results of modern distribution predicted by MaxEnt (probability parameter (≥ 0.5), Q. mongolica is mainly distributed in Liaoning, northern Hebei, and Shanxi Province. This is similar with to the present distribution. During the LGM, Q. mongolica was distributed at 30–35° N in central and southern China. There was also a high probability of distribution in Liaoning Province. The model results showed that Q. mongolica did not completely retreat to the south of the Yangtze River during LGM.

3.2.3  L  ast Glacial Maximum Distribution and Postglacial Migration of Q. mongolica According to the pollen evidence, Quercus was rarely recorded during the LGM but more than 3% of the pollen percentages appeared in the central Inner Mongolia and Gansu Province. According to the model results, the mountain areas in eastern Gansu Province had a higher distribution probability during the LGM, which was consistent with pollen evidence. In central Inner Mongolia, the distribution probability was 0.1, where was not suitable for the growth of Mongolia oak. A higher percentage of Quercus pollen appeared along the northern coast and the distribution probability by model was more than 0.6, indicating that Q. mongolica might grow in coastal areas, which may be related with better water conditions in these areas. We inferred that the Q. mongolica might also be distributed in higher latitude during LGM. Therefore, the present Q. mongolica in northern China may be the result of local diffusion during the postglacial period, rather than just long-distance migration from south to north. There is a lack of research on the phylogeography pattern of Q. mongolica, and there is still no conclusive evidence on the exact location of refugia during the LGM and the migration route after the LGM.

3.3  Glacial Refugia and the Postglacial Migration of Betula platyphylla and Betula…

51

3.3  G  lacial Refugia and the Postglacial Migration of Betula platyphylla and Betula dahurica 3.3.1  Distribution Changes Based on Pollen Data Analysis Based on the study of palynology in northern China, Betula pollen has extra representation (Li et al. 2000). The percentages of Betula pollen in the forest and meadow zones reached 30–50%, while the percentages of other samples reached 5–20%. Birch forest was considered to exist when the percentage of Betula pollen was more than 30% (Liu et al. 2002). Cao et al. (2015) selected 6% as the threshold for the > 0.7 presence probability, and a Betula pollen percentage more than 25% indicated that Betula was dominant in the field. Ren and Beug (2002) argued that a pollen percentage more than 10% represented the existence of Betula, while a pollen percentage more than 40% can be considered local dominance. In Europe, Lisitsyna et al. (2011) summarized previous studies showing thresholds of 10–70% for sedimentary pollen and 4.3–25% for modern pollen. Ralska-Jasiewiczowa et al. (2004) suggested that a pollen percentage of 20–43% could be used as a threshold. In this study, a threshold of 10% was used to indicate the existence of Betula species locally. During the LGM (24,000–20,000 cal year BP), there were relatively few pollen records (n  =  6), and only one site, which was located in northeastern China (Xingkai Lake), had a percentage greater than 10%. During the transition period between the LGM and Holocene (20,000–12,000  cal  year BP), the distribution gradually increased and began to appear in central Inner Mongolia. During the early Holocene, Betula was extensively distributed, especially in northern China, and the percentages were also relatively high in all these time spans since the LGM. Since 8000  cal  year BP, the scope of the distribution has continued to narrow, and the percentages have also decreased. In general, the distribution gradually increased after the LGM and decreased since the early Holocene, indicating that Betula was a pioneer species.

3.3.2  S  imulated Present and LGM Distribution of  B. platyphylla and B. dahurica The predicted current distribution of B. platyphylla with a logistic value ≥ 0.5 was generally similar to the observed distribution (Fang et al. 2009) along the 400 mm isohyet in northern China. During the LGM, the simulated distribution of B. platyphylla extended from 23 to 36 ° N and 100 to 112 ° E and was centered on central and southwestern China. The MaxEnt prediction results for the distribution of B. davurica (distribution logistic value ≥ 0.5) are presented, and B. davurica was mainly distributed in northeastern China, northern Hebei Province, and Inner Mongolia. During the LGM, this

52

3  Glacial Refugia and the Postglacial Migration of Dominant Tree Species…

species was mainly distributed along the coastal areas of Liaoning, Shandong, Jiangsu, and Zhejiang Province.

3.3.3  L  ast Glacial Maximum Distribution and Postglacial Migration of B. platyphylla and B. dahurica Pollen evidence showed that birch was rarely recorded in sediments during the LGM, and none > 10% of the pollen records were recorded. In the northern coastal areas, there were high pollen percentages of Betula and the distribution probability was more than 0.6, indicating the growth of Betula in the coastal areas. During the early Holocene, birch pollen had high percentages in Liaoning Province, and then there was a trend of local diffusion. Therefore, the present Betula populations in northern China may be spread from multiple refugia during the postglacial period, rather than simply migrated from south to north. There is a lack of research on the phylogeography pattern of Betula, and there is still no conclusive evidence on the location of refugia during the LGM and the migration route during the postglacial period.

3.4  Chapter Summary Through the analysis and comparison of the four main tree species in the study areas, none of the four species retreated completely to the south of the Yangtze River during the LGM, although in the phylogeography, pollen and MaxEnt were not always consistent. Especially for Pinus tabulaeformis with genetic data, the conclusion was more exact. Thus, it can be concluded that mountain areas in northern China play a refugia role to ensure species survival during the LGM, while the possibility of coastal areas and the Otindag Sandy Land as refugia needs further study. The existence of refugia during the glacial period further illustrates that tree populations migrated in various ways after the glacial period, resulting from a combination of long-distance migration from south to north and local diffusion. Various migration patterns help to correct the excessively high rate of species migration estimated based on fossil evidence. For different species, we can see that the migration of species is individual.

References Anderson, L.L., Hu, F.S., Nelson, D.M., Petit, R.J. and Paige, K.N. 2006. Ice-age endurance: DNA evidence of a white spruce refugium in Alaska. Proceedings of the National Academy of Sciences, 103, 12447–12450.

References

53

Ashcroft, M.B., Gollan, J.R., Warton, D.I. and Ramp, D. 2012. A novel approach to quantify and locate potential microrefugia using topoclimate, climate stability, and isolation from the matrix. Global Change Biology, 18, 1866–1879. Birks, H.J.B. 2015. Some reflections on the refugium concept and its terminology in historical biogeography, contemporary ecology and globalchange biology. Biodiversity, 16, 196–212. Breen, A.L., Murray, D.F. and Olson, M.S. 2012. Genetic consequences of glacial survival: the late Quaternary history of balsam poplar (Populus balsamifera L.) in North America. Journal of Biogeography, 39, 918–928. Burban, C. and Petit, R. 2003. Phylogeography of maritime pine inferred with organelle markers having contrasted inheritance. Molecular Ecology, 12, 1487–1495. Cao, X., Herzschuh, U., Ni, J., Zhao, Y. and Böhmer, T. 2015. Spatial and temporal distributions of major tree taxa in eastern continental Asia during the last 22,000 years. The Holocene, 25, 79–91. Chen, K., Abbott, R.J., Milne, R.I., Tian, X. and Liu, J. 2008. Phylogeography of Pinus tabulaeformis Carr. (Pinaceae), a dominant species of coniferous forest in northern China. Molecular Ecology, 17, 4276–4288. Cinget, B., de Lafontaine, G., Gérardi, S. and Bousquet, J. 2015. Integrating phylogeography and paleoecology to investigate the origin and dynamics of hybrid zones: insights from two widespread North American firs. Molecular Ecology, 24, 2856–2870. Collins, W.D., Bitz, C.M., Blackmon, M.L., et al. 2006. The community climate system model version 3 (CCSM3). Journal of Climate, 19, 2122–2143. de Lafontaine G., Ducousso A., Lefèvre S., Magnanou E., and Petit R.J. 2013. Stronger spatial genetic structure in recolonized areas than in refugia in the European beech. Molecular Ecology, 22, 4397–4412. de Lafontaine, G., Amasifuen Guerra, C.A., Ducousso, A., Sánchez-Goñi, M.-F. and Petit, R.J. 2014. Beyond skepticism: inferring cryptic refugia using multiple lines of evidence. New Phytologist, 204, 450–454. de Lafontaine, G., Turgeon, J. and Payette, S. 2010. Phylogeography of white spruce (Picea glauca) in eastern North America reveals contrasting ecological trajectories. Journal of Biogeography, 37, 741–751. Demesure, B., Sodzi, N. and Petit, R.J. 1995. A set of universal primers for amplification of polymorphic non-coding regions of mitochondrial and chloroplast DNA in plants. Molecular Ecology, 4, 129–134. Dobrowski, S.Z. 2011. A climatic basis for microrefugia: the influence of terrain on climate. Global Change Biology, 17, 1022–1035. Doyle, J.J. and Doyle, J.L. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin, 19, 11–15. Du, F.K., Petit, R.J. and Liu, J.Q. 2009. More introgression with less gene flow: chloroplast vs. mitochondrial DNA in the Picea asperata complex in China, and comparison with other conifers. Molecular Ecology, 18, 1396–1407. Dumolin-Lapegue, S., Pemonge, M-H. and Petit, R.J. 1997. An enlarged set of consensus primers for the study of organelle DNA in plants. Molecular Ecology, 6, 393–397. Fang, J., Wang, Z., Tang, Z. 2009. Atlas of Woody Plants in China: Distribution and Climate. Beijing: Higher Education Press. Fawcett, T. 2006. An introduction to ROC analysis. Pattern Recognition Letters, 27, 861–874. Feng, Z., Tang, L., Ma, Y., et al. 2007. Vegetation variations and associated environmental changes during marine isotope stage 3 in the western part of the Chinese Loess Plateau. Palaeogeography, Palaeoclimatology, Palaeoecology, 246, 278–291. Gavin, D.G., Fitzpatrick, M.C., Gugger, P.F., et  al. 2014. Climate refugia: joint inference from fossil records, species distribution models and phylogeography. New Phytologist, 204, 37–54. Gérardi, S., Jaramillo-Correa, J.P., Beaulieu, J., Bousquet, J. 2010. From glacial refugia to modern populations: new assemblages of organelle genomes generated by differential cytoplasmic gene flow in transcontinental black spruce. Molecular Ecology, 19, 5265–5280.

54

3  Glacial Refugia and the Postglacial Migration of Dominant Tree Species…

Hao, Q., de Lafontaine, G., Guo, D., Gu, H., Hu, F.S., Han, Y., Song, Z., Liu, H. 2018. The critical role of local refugia in postglacial colonization of Chinese pine: joint inferences from DNA analyses, pollen records, and species distribution modeling. Ecography, 41, 592–606. Harrison, S., Yu, G., Takahara, H. and Prentice, I. 2001. Palaeovegetation (Communications arising): diversity of temperate plants in east Asia. Nature, 413, 129–130. Hewitt, G. 2000. The genetic legacy of the Quaternary ice ages. Nature, 405, 907–913. Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G. and Jarvis, A. 2005. Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology, 25, 1965–1978. Hugall, A., Moritz, C., Moussalli, A. and Stanisic, J. 2002. Reconciling paleodistribution models and comparative phylogeography in the Wet Tropics rainforest land snail Gnarosophia bellendenkerensis (Brazier 1875). Proceedings of the National Academy of Sciences, 99, 6112–6117. Huntley, B. and Birks, H.J.B. 1983. An atlas of past and present pollen maps for Europe: 0–13,000 years ago. Cambridge University Press, Cambridge, UK. Jaramillo-Correa J., Bousquet J., Beaulieu J., Isabel N., Perron M. and Bouillé M. 2003. Cross-­ species amplification of mitochondrial DNA sequence-tagged-site markers in conifers: the nature of polymorphism and variation within and among species in Picea. Theoretical and Applied Genetics, 106, 1353–1367. Jiang, H. 1984. Observation on the microstructure of the pollen grains of Pinus—Discussion for the taxonomy position of Pinus Bungeana. Bulletin of Botanical Research, 4, 137–143. (in Chinese) Keller, S.R., Olson, M.S., Silim, S., Schroeder, W. and Tiffin, P. 2010. Genomic diversity, population structure, and migration following rapid range expansion in the Balsam Poplar, Populus balsamifera. Molecular Ecology, 19, 1212–1226. Li, R., Zheng, L., Zhu, G. 1990. Lake and environment in the Inner Mongolia Plateau. Beijing: Beijing Normal University Press. Li, J., Chen, F., Cook, E.R., Gou, X. and Zhang, Y. 2007. Drought reconstruction for north central China from tree rings: the value of the Palmer drought severity index. International Journal of Climatology, 27, 903–909. Li, F., Gao, S. 2008. Plant Biology. Beijing: China Forestry Publishing House. Li, Y., Zhang, X., Zhou, G., Ni, J. 2000. Quantity relationship between pollen and vegetation in north China. Chinese Science Bulletin, 45, 761–765. (in Chinses) Liepelt, S., Bialozyt, R. and Ziegenhagen, B. 2002. Wind-dispersed pollen mediates postglacial gene flow among refugia. Proceedings of the National Academy of Sciences, 99, 14590–14594. Lisitsyna, O.V., Giesecke, T., Hicks, S. 2011. Exploring pollen percentage threshold values as an indication for the regional presence of major European trees. Review of Palaeobotany and Palynology, 166, 311–324. Liu, C., Berry, P.M., Dawson, T.P., Pearson, R.G. 2005. Selecting thresholds of occurrence in the prediction of species distributions. Ecography, 28, 385–393. Liu, G., Yin, Y., Liu, H. and Hao, Q. 2013. Quantifying Regional Vegetation Cover Variability in North China during the Holocene: Implications for Climate Feedback. Plos one, 8, e71681. Liu, H., Cui, H., Pott, R. and Speier, M. 1999. The surface pollen of the woodland-steppe ecotone in southeastern Inner Mongolia, China. Review of Palaeobotany and Palynology, 105, 237–250. Liu, H., Xu, L., Cui, H. 2002. Holocene history of desertification along the woodland-steppe border in northern China. Quaternary Research, 57, 259–270. Liu, H., Yin, Y., Zhu, J., Zhao, F. and Wang, H. 2010. How did the forest respond to Holocene climate drying at the forest-steppe ecotone in northern China? Quaternary International, 227, 46–52. Liu, Y., Shi, J., Shishov, V., Vaganov, E., Yang, Y., Cai, Q., Sun, J., Wang, L. and Djanseitov, I. 2004. Reconstruction of May–July precipitation in the north Helan Mountain, Inner Mongolia since AD 1726 from tree-ring late-wood widths. Chinese Science Bulletin, 49, 405–409.

References

55

Mao, J., Wang, X. 2011. Distinct Niche Divergence Characterizes the Homoploid Hybrid Speciation of Pinus densata on the Tibetan Plateau. The American Naturalist, 177, 424–439. Milecka, K., Kupryjanowicz, M., Makohonienko, M., Okuniewska-Nowaczyk, I., Nalepka, D. 2004. Late Glacial and Holocene history of vegetation in Poland based on isopollen maps. W. Szafer Institute of Botany, Polish Academy of Sciences, Kraków, Poland. Muscarella, R., Galante, P.J., Soley-Guardia, M., Boria, R.A., Kass, J.M., Uriarte, M., Anderson, R.P., 2014. ENMeval: An R package for conducting spatially independent evaluations and estimating optimal model complexity for Maxent ecological niche models. Methods in Ecology and Evolution, 5, 1198–1205. Neale, D., Sederoff, R.R. 1989. Paternal inheritance of chloroplast DNA and maternal inheritance of mitochondrial DNA in loblolly pine. Theoretical and Applied Genetics, 77, 212–216. Neale, D.B., Wheeler, N.C., Allard, R.W. 1986. Paternal inheritance of chloroplast DNA in Douglas-fir. Canadian Journal of Forest Research, 16, 1152–1154. Ni, J., Yu, G., Harrison, S.P. and Prentice, I.C. 2010. Palaeovegetation in China during the late Quaternary: Biome reconstructions based on a global scheme of plant functional types. Palaeogeography, Palaeoclimatology, Palaeoecology, 289, 44–61. Phillips, S.J., Anderson, R.P. and Schapire, R.E. 2006. Maximum entropy modeling of species geographic distributions. Ecological Modelling, 190, 231–259. Qiu, Y., Fu, C. and Comes, H.P. 2011. Plant molecular phylogeography in China and adjacent regions: tracing the genetic imprints of Quaternary climate and environmental change in the world’s most diverse temperate flora. Molecular Phylogenetics and Evolution, 59, 225–244. Ralska-Jasiewiczowa, M., Wacnik, A., Mamakowa, K., Nalepka, D. 2004. Late Glacial and Holocene history of vegetation in Poland based on isopollen maps. W.  Szafer Institute of Botany, Polish Academy of Sciences, Kraków. Ren, G., Beug, H.J. 2002. Mapping Holocene pollen data and vegetation of China. Quaternary Science Reviews, 21, 1395–1422. Small, R.L., Ryburn, J.A., Cronn, R.C., Seelanan, T. and Wendel, J.F. 1998. The tortoise and the hare: choosing between noncoding plastome and nuclear Adh sequences for phylogeny reconstruction in a recently diverged plant group. American Journal of Botany, 85, 1301–1315. Tang, L., Li, C., An, C. and Wang, W. 2007. Vegetation history of the western loess plateau of China during the last 40 ka based on pollen record. Acta Palaeontologica Sinica, 46, 45–61. Temunović, M., Frascaria-Lacoste, N., Franjić, J., Satovic, Z. and Fernández-Manjarrés, J.F. 2013. Identifying refugia from climate change using coupled ecological and genetic data in a transitional Mediterranean-temperate tree species. Molecular Ecology, 22, 2128–2142. Tian, B., Liu, R., Wang, L., Qiu, Q., Chen, K. and Liu, J. 2009. Phylogeographic analyses suggest that a deciduous species (Ostryopsis davidiana Decne., Betulaceae) survived in northern China during the Last Glacial Maximum. Journal of Biogeography, 36, 2148–2155. Wagner, D.B. 1992. Nuclear, chloroplast, and mitochondrial DNA polymorphisms as biochemical markers in population genetic analyses of forest trees. New Forests, 6, 373–390. Warren, D.L., Seifert, S.N. 2011. Ecological niche modeling in MaxEnt: the importance of model complexity and the performance of model selection criteria. Ecological Applications, 21, 335–342. Warren, E., de Lafontaine, G., Gérardi, S., Senneville, S., Beaulieu, J., Perron, M., Jaramillo-­ Correa, J.P. and Bousquet, J.  2016. Joint inferences from cytoplasmic DNA and fossil data provide evidence for glacial vicariance and contrasted postglacial dynamics in tamarack, a transcontinental conifer. Journal of Biogeography, doi:https://doi.org/10.1111/jbi.12675. Wu, Z. 1980a. Vegetation of China. Beijing: Science Press. Xu, Q., Li, Y., Yang, X., Zheng, Z. 2005. Surface pollen assemblages of some major forest types in northern China. Quaternary Sciences, 25, 585–597. (in Chinese with English abstract) Xu, Q., Li, Y., Yang, X., Zheng, Z. 2007. Quantitative relationship between pollen and vegetation in northern China. Science in China Series D: Earth Sciences, 50, 582–599.

56

3  Glacial Refugia and the Postglacial Migration of Dominant Tree Species…

Yu, G., Chen, X., Ni, J., et al. 2000. Palaeovegetation of China: a pollen data-based synthesis for the mid-Holocene and last glacial maximum. Journal of Biogeography, 27, 635–664. Yu, G., Prentice, I.C., Harrison, S.P. and Sun, X. 1998. Pollen-based biome reconstructions for China at 0 and 6000 years. Journal of Biogeography, 25, 1055–1069. Zong, Y., Sun, P., Liu, J., Yue, X., Niu, Q. and Teng, Y. 2014. Chloroplast DNA-based genetic diversity and phylogeography of Pyrus betulaefolia (Rosaceae) in Northern China. Tree Genetics and Genomes, 10, 739–749.

Chapter 4

Effects of Vertical Migration on Local Vegetation

As mentioned in the introduction, most previous studies have focused on the North-­ South migration of forests according to climate change (Yu et al. 2000; Ni et al. 2010; Cao et al. 2015) and rarely on the vertical distribution of forests. This chapter would answer the Hypothesis 2: climate change is not only the driver of the horizontal migration of forest in forest-steppe ecotone, but also the vertical migration along with mountains. The possibility of vertical migration depends on the altitude range of forest distribution areas (altitude range = maximum altitude value of the pollen source area – minimum altitude value of the pollen source area). In order to test the above hypothesis, we used previously published sediment pollen records of eight lakes located in the forest-steppe ecotone in northern China to reconstruct temporal and spatial changes of Pinus, Quercus, and Betula. To show the role of altitude range on forest distribution as well as related arboreal pollen deposition, we adopted 63 soil surface pollen records located in this region to analyze the effect of temperature, precipitation, and topographic factors (altitude and altitude range) on forest types, especially forest occupation. A conceptual model of forest migration in mountainous forest-steppe ecotone during the Holocene was also developed.

4.1  Study methods 4.1.1  Sediment Pollen Data Collection Twenty pollen records of lake were collected in our study area (Tables 4.1, and 4.2). To ensure data quality, the pollen records used in this study were collected to meet the following criteria: (1) the location of the lake was limited to the forest-steppe ecotone; (2) the pollen record included at least 5 AMS 14C data or 1 AMS 14C data per 1000 years (AMS: Accelerator Mass Spectrometry); (3) the age range was at © Springer Nature Singapore Pte Ltd. 2018 Q. Hao, The LGM Distribution of Dominant Tree Genera in Northern China’s Forest-steppe Ecotone and Their Postglacial Migration, Springer Theses, https://doi.org/10.1007/978-981-13-2883-1_4

57

Longitude (°) 112.67

112.60 113.28 115.70 114.40 114.52 115.21 115.09

Lake name Daihai Lake

Diaojiaohaizi Huangqihai Hulun Nuur Anguli Nuur Bai Nuur Bayanchagan Ulan Nuur

41.10 40.84 41.70 41.35 41.65 41.65 41.74

Latitude (°) 40.55

0.0 4.5 1.4 2.6 1.2 3.0 1.2

MAT (°C) 5.1 350 360 426 370 350 369 375

MAP (cm) 423 2015 1265 1375 1320 1346 1355 1246

Altitude (m) 1216 864 573 414 386 339 339 275

Altitude range (m) 951 13 10 7 9 5 9 4

14

C age number 6/8 2000–12,000 0–8600 0–5700 0–11,500 6100–11,000 600–12,000 6000–8600

Age range (year) 0–12,000/0– 10,500 80 50 100 50 90 120 40

Sample resolution (year) 80/60

References Li et al. (2004) Xiao et al. (2004) Yang (2001) Hao et al. (2014) Yin et al. (2015) Liu et al. (2010) Wang et al. (2012) Jiang et al. (2006) Wang et al. (2012)

Table 4.1  Information of 8 lakes fossil pollen profiles in northern China in this study, including the geographical information, climate, altitude range (pollen source area: 20 km) and sediment pollen data information

58 4  Effects of Vertical Migration on Local Vegetation

42.62 35.54

42.37 42.95 42.07

116.76 116.76 114.48

116.62

116.82 104.52

Haoluku Liuzhouwan Yujiagou

Shandianhe

Xiaoniuchang Sujiawan

Jiangjunpaozi 117.47 Charisu 122.35 Wangxianggou 119.92

42.22

42.96 42.71 40.15

41.98

115.18

Taipusi

Latitude (°) 42.28 40.67

Longitude (°) 118.97 111.13

Sediment name Chifeng Chasuqi

394.8 402.1 415.8 430.5 431.3 441.3

−0.3 5.9

−0.2 6.7 4.1

386.4 389.7 393

−0.7 −0.2 6.4

0.4

382.4

MAP (cm) 370.9 372.6

1.6

MAT (°C) 4.6 6.4

1567 250 735

1411 1958

1261

1309 1391 855

1491

Altitude (m) 565 1011

Lake Peat Lake

Lake Lake

Lake

Lake Lake Basin

Lake

Sediment type Lake Peat

2 10 7

3 7

2

3 3 3

3

C age number 5 4

14

1000–12,000 0–5000 0–5500

3500–10,000 0–12,000

0–9000

0–11,000 200–12,000 0–12,000

0–10,000

Age range (year) 0–8000 0–10,000

700 200 190

300 290

700

200 400 450

75

Sample resolution (year) 500 70

References Xu et al. (2002) Wang et al. (1999) Huang et al. (2004) Liu et al. (2001) Liu et al. (2001) Xia et al. (2001) Wang et al. (2006) Liu et al. (2001) Feng et al. (2006) Liu et al. (2001) Li et al. (2003) Li et al. (2006)

Table 4.2  Information of 12 lakes fossil pollen profiles in northern China in this study which did not satisfy the 4 standards listed, including the geographical information, climate and sediment pollen data information

4.1 Study methods 59

60

4  Effects of Vertical Migration on Local Vegetation

least 2000 years during the Holocene; and (4) the resolution of the pollen samples was at least 150 years. Finally, eight lake sediment sequences were adopted. There were five sites with original data (Anguli Nuur, Hulun Nuur, Huangqihai, Bai Nuur, and Ulan Nuur) found by our research group. The remaining published previously pollen spectra were digitized with the Getdata Graph Digitizer. Daihai Lake had two pollen records (Table  4.1). Percentages of three dominant tree genera (Pinus, Quercus and Betula) and total arboreal pollen (AP) were recalculated at 300-year intervals in each sequence with linear interpolation of two neighboring ages. The eight lakes used in the study are all located at the southeastern edge of the Inner Mongolia Plateau and their altitude ranges from about 1200 to 2020 m. Hulun Nuur, Bayanchagan, and Ulan Nuur are close to the Otindag Sandy Plain, 70 km northwest of Bai Nuur. Diaojiaohaizi and Daihai Lake are near the Yin Mountains. Anguli Nuur and Huangqihai are located between the above-mentioned lakes (Fig. 4.1). At present, the vegetation types and patterns of the eight lakes are similar, and the forests are patchily distributed in temperate grassland.

Fig. 4.1  Location of study area at regional and local scales. (a) Map of China showing location of the study region, Dongge Cave, and the 400-mm isohyet (rectangle area as shown in detail in (b). The small black triangles around the study area indicate the soil surface pollen records (n = 63); (b) Digital Elevation Model (DEM) image of the study area showing eight lakes (Daihai, Diaojiaohaizi, Huangqihai Lake, Anguli Nuur, Bayanchagan, Hulun Nuur, Ulan Nuur, and Bai Nuur.). The dark points indicate the location of selected lakes. The gray circles indicate the pollen source area (20 km) of these lakes (Bradshaw and Webb III 1985; Tarasov et al. 2007; Williams and Jackson 2003; Xu et al. 2012). The small black triangles around the study area indicate the soil surface pollen records (n = 63). (Reproduced from with permission Hao et al. (2016). Copyright (2016) Elsevier)

4.1 Study methods

61

4.1.2  Altitude Range Calculation To analyze the altitude range, we defined a projected area of 20 km radius as the pollen source area based on previous studies (Bradshaw and Webb III 1985; Tarasov et al. 2007; Williams and Jackson 2003; Xu et al. 2012), because the pollen source area represents the area range of the reconstructed palaeovegetation. For uncertainties in the pollen source area, we also calculated altitude range using 10 km, 30 km, 40  km, and 50  km as the pollen source areas (Table  4.3). Furthermore, we used spatial analysis in ArcGIS 10.2 to obtain the mean, minimum, and maximum of altitudes in pollen source areas of the selected lakes. The altitude range (Δ altitude) is the difference value of maximum and minimum of altitude within a 20 km radius. Altitude data (DEM: Digital Elevation Model) was obtained from the website (http://www.gdem.aster.ersdac.or.jp/search.jsp). In addition, all lake sediment records in this region (n = 20; 12 of 20 lakes did not satisfy the four standards list above; Table 4.2) were divided into two groups (high altitude range group with  >  500  m altitude range and low altitude range group with   500  m; 400–500 m; 300–400 m and  500 m) and the red one indicates that of all small altitude range lakes (