Resolving seasonal variability of the Asian Monsoon at the end of the Younger Dryas from a Chinese speleothem

The record of δ¹⁸O variability in Chinese speleothems at millennial and orbital scales has been proposed to reflect changes in the strength of the Asian Summer Monsoon. This remarkable record has been constructed using conventional drilling techniques that typically incorporate material from multiple years of growth in each δ¹⁸O analysis. Here, we apply techniques that provide much higher spatial and temporal resolution in order to assess seasonal variability of δ¹⁸O in cave calcite and test hypothesized changes in seasonal monsoon strength. We use the WiscSIMS ion microprobe to measure δ¹⁸O in stalagmite sample 'BW-1' (Kulishu Cave, China) using a 10 μm spot diameter and a spot-to-spot precision of ×0.3‰ (2 s.d.). Prior conventional analysis of BW-1 identified an abrupt decrease of 2.3‰ (δ¹⁸O, VPDB) in five sequential 0.5 mm drill-spots across the end of the Younger Dryas (YD) cold period. Agreement between the ²³⁰Th chronology and the number of growth bands counted in this interval suggested that BW-1 grew continuously across the YD termination while producing annual bands, and that the termination lasted <38 years. Ion microprobe analysis across this portion of BW-1 allows us to examine three questions: 1) Does the δ¹⁸O value of the summer monsoon signal decrease at the end of the YD, or 2) does the δ¹⁸O of summer monsoon rainfall remain constant while the proportion of summer (low-δ¹⁸O) vs. winter (higher-δ¹⁸O) monsoon rainfall increases? 3) How abrupt is the shift in δ¹⁸O at the end of the YD in this location? Ion microprobe δ¹⁸O analyses (n=206) indicate that in Kulishu Cave, the δ¹⁸O of calcite precipitated during both summer and winter seasons decreased rapidly at the end of the YD. In addition, the difference in δ¹⁸O between summer and winter calcite is similar during the YD and the earliest Holocene, but increases during the transition out of the YD. These data likely reflect a combination of several factors, including changing amounts of distillation of water vapor throughout the year, an increase in cave temperature, and changes in groundwater mixing above the cave. Our high-resolution analysis shows that the duration of the abrupt ending of the YD can be characterized in two different ways. First, annual growth bands widen progressively from 25 to 320 μm across 24 years of the event before narrowing again to <20 μm. By comparison, the length of the event is only 16 years if counting between the highest and lowest values of δ¹⁸O in summer calcite growth. Both band-widths and δ¹⁸O analyses indicate that the transition may have stalled for ~4 years in the middle of the climatic shift. The variations of band-width and seasonal δ¹⁸O values observed in BW-1 at the end of the YD are strikingly similar to a comparable record from Soreq Cave, Israel. Atmospheric reorganization in the Northern Hemisphere occurred on a decadal time-scale at the end of the YD.