Branched glycerol dialkyl glycerol tetraethers as indicators for environmental parameters in a subtropical mountainous river, southern China

Highlights

• BrGDGTs are useful for soil pH and mean annual air temperature (MAAT) prediction.

• Predicted pH in soil and fluvial sediment is close to soil pH.

• BayMBT₀ Model performs well for the MAAT prediction in the study region.

• Predicted MAAT for soil (fluvial sediment) is slightly higher (lower) than measured.

• In situ production of brGDGTs impact the MAAT prediction from riverine sediments.

Abstract

Indices based on branched glycerol dialkyl glycerol tetraethers (brGDGTs) have been useful in paleo-environmental reconstructions, although their applicability in river-estuarine sediments remains unclear. This study aims to investigate the correlation between brGDGT-based indices and modern environmental parameters, and to discuss possible impact factors. In this study, we discussed the applicability of several brGDGT-based indices, with specific interests on the IR_x’ (the relative abundance of 6- vs. 5-methylated brGDGTs) and CBT’ (cyclization index of branched tetraethers based on the separation of 5- and 6-methylated brGDGTs) for pH reconstruction, and MBT’_5ME (the methylation index of branched tetraethers based on the separation of 5- and 6-methylated brGDGTs) for MAAT (mean annual air temperature) reconstruction. From the Jiulong River (JLR) catchment, we have analyzed both riverbank soil samples and riverbed surface sediment samples for brGDGTs, with the following main findings: Firstly, for pH reconstruction, the indices CBT’ and IR_x’ are useful in soil and fluvial sedimentary settings. Secondly, for MAAT reconstruction, the MBT’_5ME index is sensitive to the air temperature, and the BayMBT₀ model performs relatively better than other global or regional calibrations in the study area; By employing the BayMBT₀ model, MAAT reconstructed from soils are higher than the measured MAAT, but are generally lower than the measured MAAT when reconstructed from riverbed surface sediments. Thirdly, the disagreement between the reconstructed results from the two sedimentary archives suggests that: Signatures of brGDGTs in the aquatic environment are under strong impacts of in situ produced brGDGTs; Impacts of upper-reach brGDGTs input might have resulted in the increasingly negative offset for the reconstructed MAAT downstream; And influences of complex aquatic biogeochemical conditions on sedimentary brGDGT signals are site-dependent in this area. Finally, brGDGT-based indices are sensitive to the source of brGDGTs, e.g., soil vs aquatic in situ brGDGTs; And terrigenous brGDGTs have been exposed to a range of complex biogeochemical processes before their deposition in the estuary. Based on these findings, we suggest that at least the above impact factors need to be taken into consideration when interpreting the brGDGT signals from estuarine archives, and thus, multi-proxy studies are encouraged. The findings of this study provide a better understanding of the correlation between brGDGT-based indices and modern environmental parameters in subtropical river systems, and provide useful information for interpreting the paleo-environmental signals based on brGDGTs from fluvial-estuarine archives.

Introduction

Rapid sedimentation in fluvial-estuarine continuum provides excellent sedimentary archives for the study of land-sea interactions, coastal carbon dynamics and paleoenvironmental changes. Biomarkers have proven to be a useful tool in such settings, because of their relatively strong resistance to degradation, e.g., lipid biomarkers can persist in sediments for hundreds to millions of years (Newman et al., 2016; Yang et al., 2019). Studies show that when lipids biosynthesized, they record the general environmental conditions of the microbial community (Yang et al., 2019). Thus, indices based on lipids, e.g., branched glycerol dialkyl glycerol tetraether (brGDGTs), can provide important means to reconstruct paleo-environment parameters (Sun et al., 2019; Pei et al., 2021; Zhu et al., 2021). A thorough understanding of the correlation between indices based on modern brGDGTs and modern environmental parameters serves as a key for accurate paleo-record interpretations, which requires more researches. This is particularly important for fluvial-estuarine settings, in which the signatures of brGDGTs and their impact factors are understudied.

The brGDGTs, principal lipids in the cell membrane of bacteria, contain 4–6 methyl groups and 0–2 cyclopentane rings (Weijers et al., 2006, Weijers et al., 2010), and are ubiquitous in terrestrial and aquatic environments, e.g., soil, peat, stalagmites, lake, and ocean (Schouten et al., 2000; Weijers et al., 2007; Huguet et al., 2008; Blaga et al., 2009; Yang et al., 2011). They are thought to be produced by heterotrophic bacteria that prefer to grow in soils, peats and within the water column (Weijers et al., 2006, Weijers et al., 2010; Sinninghe-Damsté et al., 2011; De Jonge et al., 2014a; Dong et al., 2018; Cao et al., 2020). Although a full understanding on the sources of brGDGTs remains unclear, Acidobacteria has been indicated as one of their major contributors (Sinninghe-Damsté et al., 2011; Halamka et al., 2021).

Over the past decade, indices based on the composition of brGDGTs have been proven to be related to temperature and soil pH (Weijers et al., 2007; Peterse et al., 2012; Naafs et al., 2017; Dearing Crampton-Flood et al., 2020; Wang et al., 2020). The temperature and soil pH influence the distribution of brGDGTs by modulating the degree of cyclization and methylation of brGDGTs, expressed as the CBT (cyclization index of branched tetraethers) and the MBT (methylation index of branched tetraethers) (Weijers et al., 2007), respectively. Peterse et al. (2012) examined the extended global surface soil dataset and revised indices based on CBT and MBT expression to better describe the empirical relations of brGDGT composition with the mean annual air temperature (MAAT) and soil pH. Although the determination coefficient of the new transfer function for MAAT (r² = 0.59) was lower than that of original equation calculated by Weijers et al. (2007) (r² = 0.77), the extended samples made the revised indices more broadly applicable.

With the development of improved methods in liquid chromatography, De Jonge et al., 2013, De Jonge et al., 2014b managed to separate the 6-methylated brGDGTs (six novel brGDGT isomers, containing methyl groups at the α/ω 6 position) from the 5-methylalted brGDGTs. By separating the 5- and 6-methylated brGDGTs, they developed CBT’ and MBT’_5ME indices (De Jonge et al., 2013, De Jonge et al., 2014b). As soil pH was suggested to be the main controlling factor in 6-methylated brGDGTs distribution, soil pH was found to be better correlated with CBT’ than with CBT (De Jonge et al., 2014b). Furthermore, as 6-methylated brGDGTs were thought to be the main cause of the large scatter in the correlation between MBT and MAAT, the MBT’_5ME showed a better correlation with MAAT (De Jonge et al., 2014b; Yang et al., 2015). Based on previous studies, Naafs et al. (2017) refined global MBT’_5ME calibration with MAAT using an extended global soil dataset to cover all climatic zones. Recently, more indices are developed directly based on the relative abundance of 6- and 5-methylated brGDGTs (De Jonge et al., 2014a; Yang et al., 2015), e.g., isomer ratio (IR_x’). Furthermore, MAAT_mr also has been successfully applied to the reconstruction of environmental parameters (De Jonge et al., 2014b; Yang et al., 2015). A recent study by Dearing Crampton-Flood et al. (2020) further expanded the dataset from the previous calibrations with newly published soil data, and calibrated the relationship between the 5-methylated brGDGTs and MAAT using the Bayesian statistics. They found that the addition of soils from warm (>29 °C) environments has increased the upper limit of the Bayesian calibration (BayMBT). The BayMBT model has also effectively minimized the structured MAAT residuals prevalent in previous calibrations. They suggested that the MBT’_5ME index is best correlated to the average temperature of all months >0 °C, called the BayMBT₀ model. During the past decade, indices based on the separation of 5- and 6-methylated brGDGTs have been widely used in soil, peat, lake, and marine environments (Zang et al., 2018; Wang et al., 2019; Gao et al., 2020; Wang et al., 2021).

Despite their ubiquity in a variety of environmental settings, application of brGDGT-based indices in fluvial-estuarine archives are relatively limited, probably due to the complexity in both their sources (e.g., soil and in situ production of brGDGTs; De Jonge et al., 2014a) and their impact factors (e.g., aquatic biogeochemical conditions; Bechtel et al., 2010; Zell et al., 2013a). Previous studies of brGDGT-based indices were mainly conducted on river systems in tropical, temperate and polar regions, e.g., Amazon River (Zell et al., 2013a, Zell et al., 2013b) in the tropic region, Yangtze River (Zhu et al., 2011), Yellow River (Wu et al., 2014), Tagus River (Zell et al., 2014), and rivers around Qinghai Lake (Lu et al., 2016) in the temperate region, Yenisei River (De Jonge et al., 2014a), Colville River (Hanna et al., 2016), and Glomma River in the polar region (Cao et al., 2018) etc.

Rivers from subtropical region are understudied. In order to investigate the application of brGDGT-based indices in river systems from subtropical regions, the Jiulong River (JLR, southeast China) was selected. Recently, there have been studies on the characteristics of the microbial community and distribution of GDGTs in JLR (Hu et al., 2015; Li et al., 2016; Guo et al., 2018; Cheng et al., 2021), which provided foundational data for this study. For example, the brGDGTs in riverbed surface sediments were found to be primarily derived from the riverbank soil (Cheng et al., 2021), with an influence of in situ production of brGDGTs as suggested by the increasing IIIa/IIa ratio (Xiao et al., 2016). This paper aims to gain further understanding on the features of brGDGT-based indices and their applicability for environmental parameter reconstructions in river systems toward a better interpretation of paleo-records from river-estuary systems.