High seasonal variability in branched GDGT concentrations and distributions in settling particulate matter from Basin Pond, Maine, Northeastern USA

To better understand the global climate system, paleoclimate reconstructions are essential due to the lack of long-term meteorological measurements and geographical maldistribution. Branched glycerol dialkyl glycerol tetraethers (brGDGTs) are biomarkers that can be used to reconstruct past temperature. Although the brGDGT producers are not fully constrained, the MBT5ME (Methylation of Branched Tetraethers) index exhibits a significant relationship with temperature (Weijers et al., 2007; De Jonge et al., 2014), and brGDGTs preserved in lake sediments have been used to reconstruct temperature in a variety of locations, including Basin Pond (central Maine, USA) ( Miller et al., 2018). Recently, many empirical correlations between temperature and lacustrine brGDGTs have been proposed (Dang et al., 2018; Russell et al., 2018; Martinez-Sosa et al., 2021; Raberg et al., 2021; Stefanescu et al., 2021; Wang et al., 2021). However, those calibrations largely rely on the spatial variability of brGDGTs, and studies on temporal brGDGT variations are also needed, in particular to understand whether the proxy has a seasonal bias. To accomplish that, we collected 2-, or 3-week samples of settling particulate matter (SPM) from May 2018 to July 2019 in Basin Pond, aiming to better understand the connection between in situ-brGDGT production and the changing lake environment. All existing temperature calibrations were applied to the SPM samples, and all reconstructions show a similar trend, yet they do not track the seasonal temperature variations. Additionally, a pronounced brGDGT flux peak from October to November 2018 was captured when the Basin Pond experienced isothermal mixing. Similar brGDGT peaks during the lake water mixing events were also found in previous studies of mid-latitude lakes (Loomis et al., 2014; Miller et al., 2018). Such events may alter the paleoclimate interpretation of sedimentary brGDGTs, as the sedimentary brGDGT-inferred temperature may be biased to the fall months. The high variability in temperature signals and fluxes of brGDGTs, shown here and in other studies (e.g., van Bree et al., 2020), illustrates that large uncertainties remain in brGDGT paleoclimate applications.