Ice core δ¹⁸O records from low-latitude regions offer high-resolution climate data, but accurate dating remains a challenge. This study focuses on the QT No. 1 ice core from the central Tibetan Plateau, where distinct seasonal signals are lacking. The research questions are: 1) Can a reliable high-resolution chronology be established for the QT ice core despite its ambiguous seasonal signals? 2) What are the dominant climate modes reflected in the QT ice core δ¹⁸O record, and how have their frequencies changed over time? 3) What factors have contributed to the observed increase in δ¹⁸O in recent decades? Understanding these questions is crucial for reconstructing past climate variability in the region, especially regarding the influence of large-scale climate patterns like the Pacific Decadal Oscillation (PDO) and El Niño-Southern Oscillation (ENSO) on the Asian monsoon. The study's importance lies in its potential to enhance our understanding of long-term climate dynamics in a region highly sensitive to these oscillations, improving the accuracy of climate models and informing future climate projections. The Tibetan Plateau is a critical region in understanding global climate dynamics due to its sensitivity to monsoonal changes and its unique geographic location.

Previous research has extensively utilized ice core δ¹⁸O records from polar regions to reconstruct past climates, particularly over glacial-interglacial timescales. Studies on Tibetan Plateau ice cores have attempted to interpret δ¹⁸O as temperature proxies. However, recent research emphasizes the role of convective intensity, rather than local temperature, as the dominant control on precipitation δ¹⁸O. The Asian monsoon is significantly influenced by ENSO, affecting precipitation δ¹⁸O and thus influencing ice core records. Studies have demonstrated strong correlations between δ¹⁸O variations in Himalayan ice cores and monsoon indices, highlighting the influence of large-scale atmospheric circulation. Significant correlations exist between δ¹⁸O and dust in ice core records and atmospheric circulation patterns. The interannual δ¹⁸O variations in central Tibetan Plateau ice cores show an inverse correlation with the Southern Oscillation Index (SOI), indicating ENSO's influence. This correlation has also been observed in northwestern Tibetan Plateau ice cores, suggesting a broad regional impact. A major challenge in Tibetan ice core studies is precise dating, hindering the retrieval of climate signals. While annual layer counting is possible in some cores with clear seasonal signals, most inland cores lack such clarity, making dating difficult. Radioactive dating methods, such as ¹⁴C, trapped noble gases, and δ¹⁸O of trapped air, offer alternative approaches, but their resolution remains limited. This dating challenge has led to debates regarding the reliability of ice core chronologies. However, long-term correlations have been identified in other high-resolution δ¹⁸O archives like tree rings, stalagmites, corals, and giant clams, enabling the reconstruction of ENSO and other climate variabilities.

The study employed a novel approach to establish a high-resolution δ¹⁸O time series for the QT ice core. First, spectral analysis (multitaper method, MTM) of the δ¹⁸O depth series (upper 50m) was conducted to identify significant frequencies. The analysis revealed prominent peaks corresponding to annual, biennial, and ENSO modes. VMD was applied to decompose the δ¹⁸O depth series to separate these modes, particularly focusing on the ENSO mode. The selection of appropriate parameters (k value for VMD) was determined by comparing the center frequencies of the IMFs with the spectral analysis results. Once the IMFs were determined, the ENSO mode (IMF2, exhibiting the highest correlation with the original depth series) was used to establish a chronology. The peaks and troughs in the ENSO mode were aligned with those of the known ENSO index, using the biennial and annual modes as auxiliary indicators to determine the number of annual samples. This process converted the δ¹⁸O depth series into a time series. The dating method was validated using the known 1963 bomb layer and compared with earlier dating results. After establishing the chronology, the study analyzed the frequency changes, specifically focusing on PDO periodicity and shifts. Spatial correlation coefficients between the annual δ¹⁸O time series and annual SST anomalies across the Indo-Pacific region were calculated to identify the PDO signal. Wavelet analysis was used to examine the frequency changes in the reconstructed PDO signal. Finally, the recent increase in δ¹⁸O was analyzed by comparing the QT ice core δ¹⁸O with several ENSO indices and examining the frequency of El Niño and La Niña events over 30-year periods.

The study successfully established a 335-year (1677–2011 CE) high-resolution δ¹⁸O time series for the upper 50 m of the QT ice core. Spectral analysis of the δ¹⁸O depth series revealed significant frequencies corresponding to annual, biennial, and ENSO cycles. VMD effectively separated these modes, with the ENSO mode showing the strongest correlation with the original depth series. Aligning the ENSO mode with the ENSO index enabled the creation of a δ¹⁸O time series. The dating results were consistent with the known 1963 bomb layer and reasonably consistent with prior dating methods for the upper 10m. The comparison with prior dating methods revealed differences of up to 25 years, particularly in the deeper sections, highlighting the challenges in dating Tibetan ice cores. The analysis of the δ¹⁸O time series revealed strong PDO signals, with both bidecadal (25-35 years) and multidecadal (50-70 years) periodicities. Wavelet analysis indicated a shift in PDO frequency: dominance of the 50–70 year periodicity before 1900 CE, followed by increased prominence of the 25–35 year periodicity afterward. The study found a strong correlation between the reconstructed PDO and observed PDO indices, especially in the period 1900-2011 CE. The analysis of the recent increase in δ¹⁸O levels revealed a significant link with the increasing frequency of El Niño events since 1900 CE. The higher frequency of El Niño events, relative to La Niña events, is believed to be the main reason for the observed δ¹⁸O increase.

The findings address the research questions by establishing a robust chronology for the QT ice core and revealing the dominant climate modes and their frequency shifts. The successfully reconstructed PDO signal in the QT ice core demonstrates the potential of Tibetan Plateau ice cores for reconstructing Pacific climate variability over long timescales. The shift in PDO periodicity might be linked to global warming, consistent with previous studies. The findings strongly support the notion that ENSO, modulated by PDO, significantly influences the Asian monsoon and the δ¹⁸O composition of precipitation. The importance of using multiple proxy data for better ENSO reconstructions is underscored. The correlation between increasing El Niño frequency and the rise in δ¹⁸O highlights the influence of ENSO on regional climate.

This study successfully reconstructed a high-resolution 335-year record of the PDO from a Tibetan ice core, revealing a significant shift in PDO periodicity potentially linked to global warming. The findings highlight the importance of accurate ice core dating and the use of advanced signal processing techniques for extracting valuable climate information. Future research could focus on extending the time series to longer periods, improving dating accuracy, and investigating the influence of other climate factors on the QT ice core δ¹⁸O. Investigating other regional ice cores may help refine understanding of the spatial extent of these climate signals and their regional variations.

The study's limitations include the challenges associated with dating the lower portion of the ice core due to vague seasonal signals and variable annual accumulation rates. The analysis focuses primarily on the PDO's influence, while other factors influencing the ice core δ¹⁸O record are not thoroughly investigated. The study uses a single ice core; results might be strengthened by comparing with records from nearby locations.