Loess deposits in the low latitudes of East Asia reveal the ~20-kyr precipitation cycle

The East Asian Summer Monsoon (EASM) is a crucial component of the global monsoon system, influencing the ecosystem of East Asia through its abundant precipitation. Understanding the driving mechanisms of EASM precipitation is key, but inconsistencies exist among proxy indicators, leading to the "Chinese 100 kyr problem." This problem arises from the discrepancy between loess records, primarily showing a 100-kyr cycle linked to global ice volume, and stalagmite records, exhibiting a dominant 20-kyr cycle linked to precessional insolation variations. This study aims to resolve this inconsistency by investigating a high-resolution loess record from a low-latitude location within the monsoon core zone, offering a potentially clearer picture of EASM precipitation dynamics compared to mid-latitude records from the monsoon marginal zone. The research hypothesizes that precession-driven insolation changes are the primary force behind precipitation variations in this region, potentially explaining the discrepancies observed in previous studies. The importance of this study lies in its potential to refine our understanding of EASM dynamics and the mechanisms driving long-term climate change in East Asia.

Previous studies utilizing Chinese mid-latitude loess deposits from the Chinese Loess Plateau (CLP) have largely revealed a dominant 100-kyr cycle in the EASM, reflected in magnetic susceptibility (MS). Other proxies like δ¹⁸O from carbonates and land snails also exhibit this cycle, associating EASM strength with global ice volume. However, some studies, using records from the western CLP, have shown a dominant 23-kyr cycle, highlighting the precessional influence. Stalagmite δ¹⁸O records, considered proxies for monsoon intensity rather than precipitation, consistently reveal a ~20-kyr cycle, supporting the precessional forcing hypothesis. The discrepancy between these loess and stalagmite records, showing differing dominant cycles, constitutes the "Chinese 100-kyr problem." Several hypotheses have been proposed to explain this inconsistency, including the unique geographic location of the CLP, limitations of MS as a pure precipitation indicator, post-depositional smoothing effects, and the influence of moisture transport pathways. To address this problem, the present study focuses on a low-latitude loess profile in the Yangtze River valley, offering a record from the monsoon core zone, unlike previous studies largely reliant on marginal zone data.

This study utilizes a 15.9 m-thick loess profile from Madang, Pengze County, Jiangxi Province, China, located at a low latitude within the EASM core zone. 291 samples were collected and analyzed for various parameters: grain size, MS, frequency-dependent magnetic susceptibility (FDS), dithionite-citrate-bicarbonate (DCB) extractable iron (FeD), total iron (FeT), hematite content, and redness. Six optically stimulated luminescence (OSL) samples were used for age determination. Grain size analysis confirmed the eolian origin of the sediments. The MS and FDS data showed a strong positive correlation. OSL dating provided a chronological framework for the profile, extending back to ~346 ka. The Madang MS curve was correlated with that of the Luochuan profile (a well-studied CLP site), enabling the establishment of a time scale using linear interpolation and extrapolation. Wavelet and power spectrum analyses were applied to the MS and FeD/FeT time series to identify dominant cycles. The DCB method was employed for FeD extraction, with total iron (FeT) measured using ICP-OES. A detailed description of the laboratory methods for MS, FDS, FeD, and FeT measurement, along with the OSL dating protocol, including sample pretreatment, measurement procedure, and dose rate calculation, is presented in the Methods section. Statistical methods like cross-correlation analysis were used to compare the Madang record with other climate proxies and insolation data. The orbital-tuning method was used to establish a time scale for the Madang profile by correlating its magnetic susceptibility with that of the well-established Luochuan loess profile.

Wavelet and spectral analysis of the Madang loess revealed a dominant 101-kyr cycle in the MS curve, consistent with eccentricity-dominated cycles observed in CLP records. However, the FeD/FeT ratio exhibited a distinct 18-kyr cycle, closely resembling the precession cycle. This difference in dominant cycles between MS and FeD/FeT in the Madang profile contrasts with CLP loess, where both proxies show similar 100-kyr cycles. The study demonstrates that at low latitudes, FeD/FeT is a precipitation-dominated proxy, unlike in mid-latitudes (e.g., CLP) where both temperature and precipitation might significantly influence it. Correlation analyses revealed that Madang MS correlates strongly with Antarctic temperature proxies and negatively with benthic δ¹⁸O records, suggesting that MS primarily reflects temperature variations in Madang. In contrast, FeD/FeT in Madang strongly correlates with low-latitude summer solar insolation gradients, indicating a dominant influence of precessional insolation changes on precipitation. Comparisons with other records from the monsoon marginal zone (Chahanchi Lake and Xijin) indicate a synchronous 20-kyr cycle in precipitation across the broader monsoon region. This is further supported by climate simulations showing a gradual increase in the importance of the 20-kyr cycle in simulated precipitation from North to South China. However, a comparison with reconstructed local seawater oxygen isotopes (δ¹⁸Osw) from the East China Sea suggests a lack of significant precessional cycle in runoff and precipitation variations in the Yangtze River region, indicating a potential mismatch between the Madang FeD/FeT and this specific proxy. The study infers four climate modes for Madang based on the differing cycles of temperature (MS) and precipitation (FeD/FeT): warm-dry, warm-wet, cold-dry, and cold-wet, correlating with different stages of the orbital cycles.

The contrasting cycles observed in MS and FeD/FeT at Madang help address the "Chinese 100-kyr problem." The findings suggest that at low latitudes, where precipitation is abundant, FeD/FeT serves as a robust proxy for precipitation, showing the dominant influence of precession-driven insolation changes. In contrast, MS at Madang may be more sensitive to temperature variations due to redox processes influenced by soil moisture and evapotranspiration. The different responses of MS and FeD/FeT to climate factors in Madang, compared to the CLP, highlight the importance of considering regional variations in proxy responses when reconstructing paleoclimate. This study indicates that while the 100-kyr cycle might dominate temperature changes across the EASM region, the 20-kyr cycle, driven by precession, is the primary driver of precipitation variability, particularly in low-latitude areas. This supports the notion that the "Chinese 100 kyr problem" might stem from the different climate signals captured by proxies in different geographical settings within the EASM zone.

This study's high-resolution loess record from Madang provides compelling evidence that precession-driven insolation variations are the primary driver of EASM precipitation changes at low latitudes, effectively addressing a key aspect of the "Chinese 100-kyr problem." The distinct responses of MS and FeD/FeT to climate factors at Madang underscore the importance of considering regional variations in proxy behavior. Future research could focus on integrating this low-latitude record with other proxies from various regions to develop a more comprehensive understanding of EASM dynamics and their interactions with global climate systems. Further investigations into the exact mechanisms governing the relationship between MS, temperature and precipitation under various rainfall regimes would also be valuable.

The orbital tuning method used to establish the Madang timescale relies on correlation with the Luochuan profile, which itself is subject to uncertainties. Linear interpolation and extrapolation may introduce errors into the time series analysis. The study focuses primarily on the Madang site; therefore, the generalizability of the findings to other low-latitude regions within the EASM zone requires further investigation. The relatively weak correlation between Madang FeD/FeT and stalagmite δ¹⁸O records highlights potential limitations in directly comparing these proxies due to spatial and temporal variations in precipitation and the influence of local environmental factors on chemical weathering.