Figure 2. Age-depth plot for the Maar Lake Huguangyan. The calibrated AMS14C dates are from Wang et al., (2016) (red) and Chu et al., (2017) (blue).
2.3 GDGTs analysis
This study builds upon previous work of Huguangyan Maar Lake (Chu et al., 2017). For GDGTs analysis, freeze-dried samples (~1.5 g) were extracted using a dichloromethane (DCM): methanol (9:1, v/v) mixture with an accelerated solvent extractor (ASE 350) at 120 ℃ and 1500 psi for two cycles. The extracts were dried under N2 and separated into apolar and polar fractions over an activated Al2O3 column using hexane: DCM (9:1, v/v) and DCM: methanol (1:1, v/v) as the respective eluents.
The branched GDGTs were analyzed using a Shimadzu high-performance liquid chromatography triple quadrupole mass spectrometry system (HPLC-MS), with an autosampler and Analyst software (modified by Hopmans et al., 2004; Weijers et al., 2007). Separation was achieved using a Grace Prevail Cyano column (150 mm × 2.1 mm; 3 μm); ion scanning was performed in a single ion monitoring mode at m/z 1050, 1048, 1046, 1036, 1034, 1032, 1022, 1020 and 1018. The brGDGTs were quantified using a C46 internal standard.
MAAT from brGDGT for the Huguangyan Maar Lake was obtained using the degree of methylation (MBT) and the cyclisation ration of branched tetraether (CBT) according to the proxy calibration for the lacustrine sediment from China and Nepal (Sun et al., 2011):
MAAT=3.949 – 5.593 × CBT + 38.213 × MBT (n=100, r2=0.73)
3 Result and Discussion
3.1 Reconstruction temperature vs. instrumental data and historical documentary.
All proxy data are indirect measurements of climate change, and therefore it is necessary to calibrate or validate them against instrumental or other independent data(Jones & Mann, 2004). Hence, we compared the brGDGTs-based MAAT record from Huguangyan Maar Lake with several other independent climate datasets: instrumental annual mean air temperature data from Zhanjiang meteorological station, ~15 km from Huguangyan Lake; historical documents from 140 local chronicles and 14 monographs from southern China (south of 23.5ºN), and a synthetic temperature reconstruction for China (Figure 3c).
The reconstructed MAAT is 22.5℃ for the period of 2010 CE–1951 CE (based on 137Cs-dating of the uppermost part of the sediment column). This result is close to the instrumental annual mean air temperature (23.2℃) for the same period obtained from nearby Zhanjiang meteorological station, however, there is a 0.7℃ offset between the instrumental data and the reconstructed temperatures. Although temperatures in this tropical lake are favorable for biological activity throughout the year, the brGDGTs-based temperature is slightly biased to winter and spring temperatures because of the turnover of the water column, which causes nutrient-rich bottom water to reach the surface and support the growth of aquatic organisms. This was confirmed by the seasonal flux of brGDGTs estimated from sediment traps (Hu et al., 2016), together with monthly observations of planktonic diatoms (Wang et al., 2012). The occurrence of lower temperatures during the 1970s is confirmed by both instrumental data for 1921 CE–1939 CE and 1951 CE–2010 CE, as well as by the MAAT reconstruction (Figure 3a).
Historical documents have often been used for paleoclimatic reconstruction. In tropical China, historical documents are widely available and there are hundreds of local chronicles. Direct descriptions using terms such as “rivers frozen” and “snow, sleet and frost” in tropical China (south of 23º30’ N) are interpreted as clear evidence of cold events. These abnormal phenomena are related to cold surges (cold waves). Meteorologically, a cold surge has been defined as a very large temperature decrease (exceeding 8℃) within 24 hours, and they are proposed as a surrogate for winter monsoon strength in China (Ray et al., 1991). Cold surges have a relatively short duration, often less than one week. We used historical documents to produce a compilation of evidence of “rivers frozen, snow, sleet and frost in the tropical plains” (south of 23.5º N; see Supplementary Table S1). The sources are from ”Natural disasters in historical documents in Guangdong province” (including Hainan province) and ”Natural disasters in historical documents in Guangxi province” (Institute of Literary and History, Guangdong Province, unpublished data), which consist of 140 local chronicles and 14 monographs.