Temperature variations during the past 20 ka at Huguangyan Maar Lake in tropical China and dynamic link
Qi Li1,2, Qing Sun3, Manman Xie3, Yuan Ling4, Zeyang Zhu1, Qingzeng Zhu1, Nan Zhan3, Guoqiang Chu1,5,6
1Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China.
2 University of Chinese Academy of Sciences, Beijing 100049, China.
3 National Research Center of Geoanalysis, Beijing 100037, China.
4 Institute of Geology, Chinese Academy of Geological Sciences, Beijing 100037, China.
5 CAS Center for Excellence in Life and Paleoenvironment, Beijing 100044, China.
6 Innovation Academy for Earth Science, Chinese Academy of Sciences, Beijing 100029, China.
Corresponding author:
G. Chu (chuguoqiang@mail.igcas.ac.cn) and Q. Sun (sunqing1616@yahoo.com)
Key Points:
Abstract
Discrepancies exist in global temperature evolution from the Last Glacial Maximum to the present between model simulations and proxy reconstruction. This debate is critical for understanding and evaluating current global warming on a longer timescale. Here we report a branched GDGTs-based temperature reconstruction from the sediments of Huguangyan Maar Lake in southeast China and validate it using historical documentary evidence and instrumental data. The reconstructed mean annual air temperature (MAAT) indicates distinct changes during the last deglaciation (Oldest Dryas, Bølling-Allerød, Younger Dryas). During the Holocene, temperatures gradually increased from the end of the Younger Dryas to ~7.0 ka BP, followed by a decrease in recent decades. However, our terrestrial temperature record differs with model simulations and proxy sea surface temperature records of the Holocene. We conclude that ice volume or ice sheet is the most prominent forcing that controlled the regional temperature evolution from the Last Glacial Maximum to the beginning of the middle Holocene; while the temperature variations during the middle and late Holocene were mainly regulated by several possible factors, such as oceanic and atmospheric circulation, and external drivers (solar and volcanic activity).
1 Introduction
There is an ongoing debate about the Holocene temperature record of the extratropical Northern Hemisphere. Proxy records show a general long-term cooling trend since the early Holocene (Marcott et al., 2013), while model simulations indicate a warming trend over the past ~12,000 years (Liu et al., 2014). Models play an important role in understanding how climate systems respond to various forcing factors and boundary conditions. However, many dynamic processes are not well computed or weighted in the model simulations, such as external forcing factors (solar irradiance and explosive volcanism), and the internal variability of the climate system (e.g., ENSO, PDO, AO, and the monsoon). Moreover, proxy-based temperature reconstructions are often of low resolution and have relatively large uncertainties in proxy calibration and dating, as well as regional or seasonal biases. Hence, more regional proxy-based paleoclimate time series are needed to verify the climatic record of the Holocene and to understand the differences between model simulations and paleoclimate reconstructions.
In this study, we present a high resolution branched glycerol dialkyl glycerol tetraethers (brGDGTs) membrane lipids-based temperature reconstruction (0-10 ka BP) combined with previously published data (10-20 ka BP) from the sediments of Huguangyan Maar Lake (21º9´N, 110º17´E) in tropical China (Chu et al., 2017), with the aim of obtaining a regional terrestrial temperature record and to understand its dynamic links.
Our paleotemperature proxy is based on brGDGTs, which comprise two dialkyl chains with different amounts of methyl and cyclopentane moieties (Damsté et al., 2009; Hopmans et al., 2004; Naafs et al., 2017; Sanchi et al., 2014; Schouten et al., 2000; Schouten et al., 2013; Sun et al., 2011; Tierney & Russell, 2009; Weijers et al., 2007). The physical and biological mechanisms of the temperature sensitivity of brGDGTs could be due to their membrane components maintaining membrane fluidity via methyl and cyclopentane moieties (Damsté et al., 2002; Huguet et al., 2007; Weijers et al., 2007). The methylation index of branched tetraethers (MBT) and the cyclization ratio of branched tetraethers (CBT) have previously been used to reconstruct terrestrial paleotemperatures from lacustrine sediments, soils, and peat sequences (Ding et al., 2015; Naafs et al., 2017; Peterse et al., 2012; Weijers et al., 2007; Yang et al., 2014).
Numerous studies have confirmed that brGDGTs-based indices from lacustrine sediments can be used to reconstruct terrestrial paleotemperatures (De Jonge et al., 2014; Hu et al., 2016; Kaiser et al., 2015; Loomis et al., 2012; Martin et al., 2020; Naafs et al., 2017; Pearson et al., 2011; Russell et al., 2018; Sun et al., 2011; Tian et al., 2019; Tierney et al., 2010; Weijers et al., 2007; Zink et al., 2016). However, several interpretational uncertainties remain, such as regarding the relative contributions of aquatic sources and soil sources to brGDGTs because the calibration functions from soils and lakes are quite different. Huguangyan Lake is a small, hydrologically-closed lake with no inflows and a small watershed area, and therefore the sedimentary organic matter originates mainly in the water column (Chu et al., 2017; Hu et al., 2016). The within-lake origin of the sedimentary organic matter, which is also supported by the relatively low sedimentary TOC/N ratios since ~20 ka BP (Chu et al., 2002), substantially reduces this source of uncertainty in paleotemperature reconstruction. Previous brGDGTs-based MAAT reconstructions during the last deglaciation from the sediments of Huguangyan Maar Lake demonstrated a distinctive pattern of temperature changes from during the Oldest Dryas, Bølling-Allerød, Younger Dryas, and the onset of the Holocene (Chu et al., 2017).
2 Materials and Methods
2.1 Study site
Maar lakes are recognized as ideal sites for preserving high-resolution sedimentary archives because they are closed basins with a relatively simple hydrological system and they provide continuous sedimentary sequences (Yancheva et al., 2007). Huguangyan Maar Lake is a closed basin without stream inputs located in the Leizhou Peninsula in the tropical region of South China. The surface area is 2.3 km2, the maximum water depth is 22 m, and the watershed area is 3.2 km2 (Figure 1) (Chu et al., 2002; Yancheva et al., 2007). The area has a mean annual air temperature of 23.4℃ and a temperature difference between winter and summer of 11.9℃ (1964–2004; data from Zhanjiang meteorological station). Overlapping piston cores were collected from near the center of the lake in a water depth of 14 m. The cores were sliced at a 1-cm interval and then freeze dried for GDGTs extraction.