A seasonally resolved stalagmite δ18O record indicates the regional activity of tropical cyclones in Southeast China

A seasonally resolved stalagmite δ18O record indicates the regional activity of tropical cyclones in Southeast China | npj Climate and Atmospheric Science

Abstract

Identifying tropical cyclone (TC) signatures in paleoclimate records enhances our understanding of long-term TC activity trends and the climatic factors influencing TC evolution. Stalagmites are considered promising archives for recording TC activity. However, despite the western North Pacific being the most TC-active ocean basin globally, it lacks stalagmite-based TC reconstructions. Here, we present a seasonally resolved stalagmite δ18O record from XRY cave in Southeast China, covering the period from 1951 to 2018 CE, to identify annual signals of strong TC activity. We propose that the minimum seasonal XRY δ18O value of each year can reconstruct regional TC activity, achieving an identification rate of 86% for strong TC years in study area. This demonstrates the feasibility of using stalagmites for TC reconstruction in Southeast China. Moreover, our research shows that inland stalagmites can still capture TC activity signals, which will promote the use of stalagmites in obtaining long-term records of post-landfall TC activity and inland impacts.

Introduction

Tropical cyclones (TCs) cause considerable social and economic repercussions. Between 2002 and 2019 1, an estimated 560 million people were annually affected by TCs, with consequences extending beyond immediate economic and human losses. These consequences encompass potential threats to food security, human health, and biodiversity 2, 3, 4. Global warming may further amplify the harmfulness of TCs in the future. The findings from climate model projections suggest a prospective rise in the frequency and intensity of TCs 5, 6. However, instrumental records are insufficiently long to ascertain whether observed trends in TC activity fall within the spectrum of long-term natural variability or are linked to climate change 7, 8. Consequently, the development of high-resolution and longer proxy records of TCs can facilitate understanding of the long-term trends in TC activity. Additionally, it can enhance our comprehension of multiple climatic factors influencing changes in TC genesis, such as spatial changes in sea surface temperature 9 or large-scale circulation 10, and thus lead to more accurate projections of future changes in TC activity.

Currently, coastal sediment-based proxies are the most common proxies for TC reconstructions, offering TC records spanning centuries to millennia 11, 12, 13, 14, 15. However, the sediment-based proxies lack high-resolution records (such as annual resolutions) and are largely dependent on coastal geomorphology, hydrodynamic conditions, and TC characteristics 11, 16. Additionally, certain studies have attempted TC reconstructions using marine archives such as tridacna 17 and corals 18, 19. Nevertheless, these geological records are limited by their spatial distribution (oceanic or coastal zone), which makes it difficult to provide information on inland TC activity. Heavy rainfall and subsequent flooding induced by TCs can also lead to widespread damage in inland areas 20, 21. Furthermore, the longer the distance a TC travels, the broader its potential impact in inland areas 22. Recent studies highlighted a significant slowdown in the decay of landfalling TCs in recent decades, particularly in North Atlantic 23, 24 and western North Pacific 25, increasing the duration and distance of TCs over land and resulting in increased precipitation 26. Therefore, TC reconstructions from terrestrial archives (e.g., tree rings 27, 28 and stalagmites 29, 30) are critical to systematically understanding long-term trends in TC activity and mitigating losses in coastal and inland areas.

Stalagmites, characterized by their wide spatial distribution, precise dating, and uninterrupted records, are vital geological archives in paleoclimate research 31. Stalagmites that experience rapid growth in optimal conditions can record TC activity 29, 32, 33, given the distinct isotopic signature of TC precipitation compared to normal rain 34, 35, 36. Thus, stalagmites are potentially good archives for long-term, high-resolution TC reconstructions and are expected to provide detailed insights into TC activity inland. However, stalagmite-based TC reconstructions worldwide are scarce, with most previous studies concentrated on the western North Atlantic 29, 30, 32, 33, 37, 38. Besides, these caves are within 100 km of the coastline, reflecting mainly coastal TC activity. Further evidence is needed to ascertain whether stalagmites can indeed capture TC signals in inland areas.

Southeast China (SEC) stands out as one of the world’s most TC-prone regions, densely populated and facing significant TC-related risks 39, 40. In SEC, rich cave resources have supported many stalagmite studies 31. However, although cave monitoring studies discuss the possibility of reconstructing TC activity from stalagmites 41, 42, to date, there are no such reconstructions available. Additionally, unlike the western North Atlantic, controlling factors of stalagmite δ18O in SEC are complex. Regional precipitation, seasonal precipitation ratio, and moisture source variation all influence stalagmite δ18O on different timescales 43, 44. Whether potential TC signals in stalagmites will be obscured under such complex conditions remains unclear.

In this study, a stalagmite (XRY24) characterized by annual δ18O laminae spanning the years 1951 to 2018 CE was obtained from XRY Cave (Fig. 1a), located approximately 300 km inland from the coastline in SEC. Through comparison with meteorological data and simulated data from an isotope-incorporated Global Spectral Model version 2 (IsoGSM2), we examined the paleoclimatic significance of XRY24 δ18O (XRY δ18O). Notably, we investigated its capacity to capture annual signals of regional TC activity, aiming to broaden the spatial scale for stalagmite-based TC reconstruction and establish a foundation for utilizing stalagmites to reconstruct long-term, high-resolution TC activity in SEC.

a The star indicates the location of XRY cave; Gray lines represent TC tracks entering the buffer zone of the XRY cave (red circle) during 1952–2017 CE. The base map shows the windspeed-weighted TC frequency density during the same period. b Monthly mean precipitation, ACE, and δ18Op in the XRY region. Reanalysis monthly mean δ18Op is from Wang et al. 108.

Results

Chronology and oxygen isotopic results

The 230Th age (2019 ± 3 CE) of the upmost subsample agrees with the collection time (9/2018 CE) of the actively growing stalagmite, so we assume the top age of XRY24 to be 2018.67 CE. There are clear cycles in the XRY δ18O sequence consistent with observation and simulation data that show the regional precipitation δ18O (δ18Op) has a prominent annual cycle signal (Fig. 2b). We layer counted 67.7 ± 2.5 cycles in the XRY δ18O sequence. The layer counting result is generally consistent with the COPRA-based age model (Supplementary Fig. 1a). Considering that the counting result has less error, we used XRY δ18O cycles count to build the age model, which indicates the stalagmite grew from 1951 to 2018 CE.

a Comparison between XRY δ18O (this study) and δ18O from NB cave 56. The gray shaded bar represents the annual cycles of (b). b Annual cycle of XRY δ18O compared with the δ18Op from Guangzhou 109(solid red line) and IsoGSM2 (dotted red line). The dots mark the most positive/negative δ18O value in each annual cycle.

We used the Hendy test 45 and replication test method 46 to determine whether XRY24 is isotopic equilibrium. Within the same growth layer, XRY δ18O exhibit minimal variation from the center to the edge (Supplementary Fig. 2a), and there is no significant correlation between XRY δ18O and δ13C (Supplementary Fig. 2b). Besides, the annual XRY δ18O sequence is of high consistency with that of the NB cave 30 km away (r = 0.58, p < 0.05, Fig. 2a). Therefore, we conclude that XRY24 was deposited under equilibrium fractionation condition, and XRY δ18O can be used to further interpret the hydroclimatic changes in the study area.

The XRY δ18O records vary from −3.15‰ to −7.16‰ (average −5.24‰) with a high temporal resolution of 0.24 years. We used linear interpolation (sampling rate = 0.25 years) to convert XRY δ18O records into evenly spaced data for the following analysis.

Regulation of precipitation seasonality and ENSO on XRY δ18O

Under isotopic equilibrium conditions, stalagmite δ18O is mainly controlled by the δ18O of drip water and cave temperature, which typically aligns with the local mean annual temperature 31. The oxygen isotope fractionation coefficient of water-carbonate controlled by the temperature is about −0.23‰/°C 47. In recent ~60 years, the maximum variation in XRY mean annual temperature is about 2.1 °C (Supplementary Fig. 1c), which at most contribute 0.5‰ to the observed range in XRY δ18O values over this period (~2.3‰). Therefore, the XRY δ18O variability is predominantly influenced by drip water δ18O and can be used to infer changes in δ18Op.

Previous studies have shown that, on the interannual timescales, the δ18Op in SEC is mainly related to the seasonality of precipitation 43, 44. There is a significant and negative correlation (r = -0.35, p < 0.05) between XRY δ18O and the precipitation ratio of East Asian summer monsoon (EASM) to non-summer monsoon (NSM) (EASM/NSM ratio) (Fig. 3c), indicating that the seasonal variation in precipitation can affect XRY δ18O. This is mainly due to the differences in δ18Op (positive in NSM and negative in EASM) between EASM and NSM seasons, further controlled by seasonal differences in the moisture transport pathway (moisture source & transport distance) and upstream convection 48, 49, 50, 51. NSM precipitation mainly comes from the adjacent Western Pacific, and during EASM, moisture increases from the farther Indian Ocean and the trans-equatorial Pacific (Supplementary Fig. 3). In Guangzhou, the contribution of remote moisture sources (Indian Ocean, equatorial Pacific) increases from 5% of NSM to 30% of EASM 44. The moisture transported over long distances has a stronger rainout effect (makes the δ18Op negative). Therefore, the increasing EASM/NSM ratio means that the EASM precipitation increases or NSM precipitation decreases, resulting in an increase in the proportion of long-distance moisture transported in the annual precipitation, and the δ18Op becomes more negative. Numerous studies have emphasized the significant influence of upstream convection on δ18Op at low and middle latitudes 34, 52, 53. In EASM, the strong convection in the tropical moisture source region increases local precipitation, making the δ18O of moisture transported outward from the source region more negative. Qiu et al. 51 believe that upstream convective and precipitation processes play a dominant role in driving δ18Op variability, whereas variability in moisture sources does not emerge as a major factor in δ18Op variability on the interannual timescale. Outgoing longwave radiation (OLR) reflects the convection at low latitudes, and the lower OLR value indicates that convection is more intense 54. We used IsoGSM2 to obtain the spatial correlation between annual δ18Op in the study area and OLR. There are significant positive correlations (p < 0.05) between the δ18Op and the OLR index in many moisture source regions, such as the Eastern Indian Ocean, South China Sea, and Equatorial Pacific, indicating that convection contributes to the variability of δ18Op (Supplementary Fig. 4).

a Seasonal and annual XRY δ18O. b Comparison of annual XRY δ18O with MEI. c Comparison of annual XRY δ18O with the EASM/NSM ratio. d Comparison of interannual XRY δ18O with ACE obtained by CEEMDAN decomposition. Gray columns depict the TC active periods. * means confidence levels of 0.1 and ** means 0.05.

El Niño-Southern Oscillation (ENSO) is considered to affect the δ18Op and stalagmite δ18O in SEC mainly through convection 55, 56 and the West Pacific Subtropical High (WPSH) 43, 57, 58. The Multivariate ENSO Index (MEI) is negatively correlated with the EASM/NSM ratio (r = −0.37, p < 0.01). XRY δ18O positively correlates with MEI (r = 0.23, p = 0.07, Fig. 3b). Furthermore, the short periods (1.8, 3, and 7.5 years) in XRY δ18O timeseries are similar to the 2-7 years cycles of ENSO (Supplementary Fig. 5). These results indicate that the δ18Op and XRY δ18O are controlled by precipitation seasonality regulated by ENSO on the interannual timescales. A lower (higher) EASM/NSM ratio occurs during El Niño (La Niña) years; because NSM δ18Op is more positive than EASM δ18Op, the δ18Op of the whole year increases (decreases), which is then reflected in XRY δ18O.

Two datasets, CRU_TS_4.07 and IsoGSM2, are used to verify our conclusion. During the NSM season, there is a significant positive correlation between MEI and the precipitation over most of SEC (Supplementary Fig. 6a), with El Niño years experiencing more precipitation than La Niña years (Supplementary Fig. 6c). The spatial pattern during the EASM season is more ambiguous, with no significant correlation and no apparent difference in precipitation between El Niño and La Niña years (Supplementary Fig. 6b, d). Furthermore, the spatial pattern of simulated δ18Op shows a significant positive correlation between MEI and δ18Op in SEC, with mean annual δ18Op higher in El Niño years than in La Niña years (Supplementary Fig. 7a, b). These spatial patterns support the regulating effect of ENSO on XRY δ18O.

The influence of TC activity on XRY δ18O

Observational data show that the minimum value of TC δ18Op can range from ~−12‰ to −20‰, which is significantly more negative (4‰–10‰ lower) than normal monsoon precipitation 59, 60, 61, 62. This is mainly related to the moisture source and the rain shield effect. The former means that TC brings more distant moisture to land than monsoon precipitation, resulting in lower δ18Op values 41, 63. The mechanism of the rain shield effect is as follows: the strong convection inside the rain shield causes moisture to condense and fall rapidly in the oversaturated environment, and because the kinetic fractionation is weak, the δ18Op becomes extremely negative 36, 60, 64, 65.

Because XRY δ18O is mainly dominated by δ18Op, we propose that XRY δ18O is affected by regional TC activity. The mean annual XRY δ18O and accumulated cyclone energy index (ACE) demonstrate a weak correlation (r = −0.17, p < 0.1); however, a notable negative correlation emerges between them at 5-9 years scale (r = −0.47, p < 0.05, Fig. 3d). These results suggest that XRY δ18O, to some extent, records the signal of regional TC activity. When the TC activity intensifies, the strong negative TC δ18Op signal causes XRY δ18O to be more negative.

TC signals in SEC δ18Op

Hong Kong (HK, about 300 km from XRY cave) exhibits similar annual characteristics in both δ18Op and TC activity compared to XRY cave (Fig. 1b and Supplementary Fig. 8). We used long-term, continuous δ18Op records from HK to verify the relationship between TC activity and δ18Op in SEC. The data show that TCs primarily impact HK during a three-month window (July–September), coinciding with the period of minimum δ18Op values (Supplementary Fig. 4a). Based on this observation, we constructed a minimum seasonal δ18O value (δ18Omin) for each year in HK for comparison between δ18Op and ACE in HK (Fig. 4b). Both mean annual δ18Op and δ18Omin show negative correlations with ACE. However, the correlation between ACE and δ18Omin (r = -0.41, p < 0.01) is statistically stronger than that between ACE and mean annual δ18Op (r = −0.24, p = 0.13). This suggests that δ18Omin is more directly influenced by the enhanced negative δ18Op signal associated with stronger TC activity, making it a potentially superior indicator for TC activity.

a Monthly mean δ18Op. The red curve represents more TC years (ACE > median), and the blue curve is fewer TC years (ACE < median). The darker part means July to September. b Curve comparison of HK ACE (red), mean annual δ18O (gray), and δ18Omin (blue). c Distribution of HK δ18Omin in different ENSO states. The lines represent the mean value under each distribution.

The combined effect of ENSO and TCs on δ18Omin

Studies have shown that ENSO can affect precipitation δ18Omin in SEC through convection and WPSH 55, 56, 57, 66. Lu et al. 56 find that the precipitation anomalies under different phases of ENSO are related to the OLR anomalies over study area. During the EASM of El Niño (La Niña) years, the positive (negative) anomalies of OLR weaken (enhance) convection and thus decrease (increases) EASM precipitation. Ruan et al. 55 propose the importance of upstream convection. During El Niño, OLR increases and precipitation decreases in the moisture source area of the South China Sea, reducing the δ18O loss of moisture, thus leading to high δ18Op in the downstream area. The variation in the intensity and location of WPSH affects the contribution of different moisture sources and precipitation seasonality, thus regulating the δ18Op 57, 66, 67. In the EASM of El Niño years, the WPSH retreated eastward, decreasing precipitation 56.

Correlation analysis reveals a significant positive correlation between MEI and HK δ18Omin (r = 0.55, p < 0.01, Supplementary Fig. 9), meaning HK δ18Omin increases (decreases) in El Niño (La Niña) years. Interestingly, the ACE exhibits a negative correlation with HK δ18Omin. Therefore, in El Niño years, the positive δ18Omin signal associated with ENSO may weaken the negative signal caused by TCs, leading to a less prominent TC signature in the δ18Omin record. Conversely, in La Niña years, the negative δ18Omin signal from TCs may be amplified, allowing for a clearer reflection of TC activity in δ18Omin. This phenomenon is evident in the HK observational data (Fig. 4c). During the La Niña period, HK δ18Omin in more TC years is the most negative (−9.97‰). The δ18Omin in more TC years during the El Niño period (−7.78‰) is significantly higher than the former, even higher than the average of more TC years (−8.36‰), and approaching the value (−7.33‰) of fewer TC years. Frappier 33 reported a similar phenomenon in the western North Atlantic, they found that residual TC water from Hurricane “Mitch” masked the stalagmite δ18O record of a major El Niño event. Our finding suggest that ENSO can impact the δ18Op signals of TC in SEC. Given that stalagmites δ18O primarily inherits δ18Op signals, caution is warranted when using stalagmite δ18O to reconstruct TC activity in SEC due to potential interference from ENSO.

Identification of TC activity by XRY δ18Omin

We use percentiles (P90 and P10) to divide the states of XRY TC activity and XRY δ18Omin: strong TC years (STYs, ACE > P90) and abnormal bias negative events (δ18Omin < P10). In XRY region, there were a total of seven years classified as STYs from 1952 to 2016 CE. Three of the seven STYs occurred during La Niña periods, two were in neutral years, and two were during El Niño periods. A total of 7 abnormal events were extracted from the same period, and six of these events correspond to STYs (Fig. 5a). The identification rate of STYs is 86% (6/7). The results (Identification number/Total number) under different ENSO states are 3/3 (La Niña), 2/2 (neutral), and 1/2 (El Niño) (Fig. 5b). The only unidentified year (1983) happened to coincide with a strong El Niño period, suggesting that the TC signals might have been masked by it. In general, the high identification rate for STYs suggests that XRY δ18Omin can effectively indicate intense TC activity.

a Comparison of prewritten XRY δ18Omin record with XRY ACE. Red fillings represent abnormal partial negative events and strong TC years, respectively. The dotted lines show the criteria determined by percentiles. The bands indicate that the negative events correspond to STYs. In the bar chart of MEI, the parts filled in gray are represented as |MEI| > 0.5 (El Niño or La Niña years). b Identification results in different ENSO states. The black border indicates the total number of STYs, and the blue fillings indicate the number of successful identifications.

Discussion

This study presents a high-resolution and precisely dated stalagmite δ18O record spanning 1951-2018 CE from XRY cave in SEC. Our analysis reveals that on the interannual timescales, XRY δ18O are primarily controlled by the precipitation seasonality modulated by ENSO, consistent with observations from other SEC stalagmites 43, 44, 55, 56, 57, 58. During STYs, the δ18O signal of TC precipitation is superimposed on the already negative background δ18Op in the EASM season, resulting in significant negative anomalies in XRY δ18O. We used XRY δ18Omin to identify STYs with a high success rate of 86%. Furthermore, our analysis of Hong Kong meteorological data revealed an interference effect of ENSO on the TC signal within δ18Op. This phenomenon is reflected in the identification results based on XRY δ18Omin (Fig. 5b). Although the limited number of STYs makes it challenging to satisfactorily confirm this pattern in XRY δ18O, we emphasize the necessity of excluding ENSO interference when using stalagmite δ18O in SEC to reconstruct TC activity accurately in the future.

The consistent annual characteristic of δ18Op and TC activity form the foundation for our identification of TC activity on an annual scale using XRY δ18Omin. We assessed global patterns of δ18Op and TC activity, dividing the TC-affected regions of the mid-to-low latitudes into six distinct parts 68: the western North Pacific (WNP), western South Pacific (WSP), eastern North Pacific (ENP), western North Atlantic (WNA), North Indian Ocean (NIO), and South Indian Ocean (SIO). Except for NIO, which exhibits a bimodal distribution of TCs, the other five regions display remarkably similar characteristics in both δ18Op and TC activity (Supplementary Fig. 10). This similarity suggests the potential of stalagmite δ18Omin to serve as a global proxy for TC reconstruction. However, this preliminary analysis requires further extensive monitoring and paleoclimate research for confirmation.

We present a stalagmite-based TC record from SEC and WNP, demonstrating the feasibility of using stalagmites for TC reconstruction in this region. WNP is the most TC-active ocean basin globally 69, and extensive research has focused on TC frequency 70, intensity 71, and activity patterns 28, 72, revealing influences from both human and natural factors 73, such as aerosols 74, greenhouse gases 75, interdecadal Pacific oscillation 76, and the subtropical high-pressure system 77. However, due to limitations of short-term observational data 7, 8 and differences in the TC datasets 78, 79, ongoing debates remain in many aspects 71, 80, 81, 82. The anticipated high-quality stalagmite TC records will offer promising insights into the long-term trends and variability of TC activity while helping elucidate the roles of different factors in regulating TC activity. Moreover, unlike previous stalagmite studies, XRY cave is located far from the coastline, which illustrates the potential for stalagmites in inland areas to record TC activity and holds promise for improving our knowledge of TC inland impacts and life cycle. This finding is important given the recent studies emphasizing the increasing impacts of TCs in inland areas 23, 24, 25, 26. These advances will be valuable for improving TC prediction models, assessing TC risk under different climate change scenarios, and developing mitigation strategies for vulnerable regions where TCs cause significant socio-economic impacts.

Methods

Definition of EASM/NSM ratio

Previous studies in SEC showed that the onset of EASM typically commences in the late stage of May, extending through September 83. However, this initiation exhibits pronounced variability on the interdecadal scale, a phenomenon primarily regulated by ENSO and Pacific Decadal Oscillation (PDO) 84, 85. During warm PDO and ENSO phases, the EASM tends to commence later, typically in June. Conversely, in the cold phases, it commences earlier, usually in the end of April to May 86. Zhang et al. 43 proposed on this basis that representing EASM precipitation in SEC should eschew fixed periods. Instead, it should be delineated as May to September for cold PDO and ENSO phases and June to September for warm PDO and ENSO phases on the interdecadal scale. In this study, we build upon prior findings and use the precipitation from May to September to represent the EASM precipitation for 1957–1976 CE and 2000–2017 CE and precipitation from June to September for 1977–1999 CE (Supplementary Fig. 11).

Chronology

Two methods are used to establish the chronology model: U-Th dating and δ18O lamination counting. Five powder subsamples were collected with a handheld carbide dental drill, paralleling the growth planes of the stalagmite. Dates were measured on a Multi-Collector Inductively Coupled Plasma Mass Spectrometry (MC-ICP-MS, Neptune plus) in the Institute of Global Environmental Change, Xi’an Jiaotong University. The chemical procedure used to separate uranium and thorium was similar to that described in Edwards et al. 87. The instrumentation, standardization, and half-lives are reported in Cheng et al. 88, 89. Age model was constructed by using the COPRA program 90.

The annual laminations in stalagmites can provide accurate age indications for paleoclimate proxies measured within the stalagmite 91. We use XRY δ18O lamination to build the chronology model. To avoid the accuracy of the results due to personal subjectivity or visual fatigue during the counting process, different people were counted six times successively, and the chronology model was obtained after averaging.

Isotopic analyses

We drilled 283 subsamples for δ18O analysis along the growth planes of XRY24 at an average interval of 0.2 mm. The subsamples were measured by Delta V Advantage with a gas bench at the School of Resources and Environment, Linyi University. A laboratory standard SD2 was analyzed every eight subsamples to check data reproducibility. The precision of measurement is <0.1‰ with 2σ analytical errors. All stable isotope compositions are reported per mil relative to the Vienna Pee Dee Belemnite (VPDB) standards.

Accumulated cyclone energy index (ACE)

We use ACE as an indicator of the strength of regional TC activity. ACE can well reflect the comprehensive impact of TC activity—their frequency, their intensity, and their longevity, and has been widely used in TC-related studies 92, 93, 94. The ACE is the cumulative sum of the squared maximum winds observed every 6 h for a TC that reaches Tropical Storm status 92:

where Vmax2 is the maximum wind speed in meters per second at 6-h intervals.

The 6-hourly track and maximum windspeed data of TC (1952–2017 CE) were sourced from the CMA Tropical Cyclone Best Track Dataset 95, 96. This database covers TCs that develop over the western North Pacific. The basin is to the north of the equator and to the west of 180°E, and includes the South China Sea. Referring to previous studies in SEC, we chose 300 km as the radius of buffer zones 36, 97. Only TCs passing within the radius from XRY cave were used to compute annual ACE for each year from 1952 to 2017 CE. The results serve to quantify the annual influence of TCs on the study area. The ACE for HK was likewise determined using this method.

Correlation analysis method

We use MATLAB function “corr2” to calculate the linear Pearson correlation coefficient r and p-value between two records, where autocorrelation (i.e., serial dependence) has been considered. The p-value accounts for autocorrelation in both records by using AR(1) processes as the null hypothesis rather than white noise as in the standard “corr” function. Additionally, linear trends were removed from all records subjected to correlation analysis in this paper.

Signal decomposition

We employ the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) 98, 99 for signal decomposition. CEEMDAN is an enhanced version of the Empirical Mode Decomposition (EMD) algorithm that effectively decomposes signals by injecting adaptive noise into the original signal and updating the residual signal and intrinsic mode functions during each iteration. This approach effectively avoids mode mixing and end-effect issues.

Multivariate ENSO index (MEI)

ENSO is the Earth’s strongest climate fluctuation on inter-annual time scales and has global impacts although originating in the tropical Pacific. MEI is considered as a representative since it is the first principal component of six main observed variables over the tropical Pacific: sea level pressure, zonal and meridional components of the surface wind, sea surface temperature, surface air temperature, and cloudiness of the sky 100, 101. The MEI monthly values are standardized with respect to a 1950–1993 reference period and are expressed as fractions of standard deviation for which it has a total mean equal to 0 and a standard deviation equal to 1 100, 102. MEI < −0.5 is defined as the La Niña (cold) phase and MEI > 0.5 as the El Niño (warm) phase ( https://www.psl.noaa.gov/enso/mei/#overview). In this study, the mean annual MEI > 0.5 (<−0.5) is used to define El Niño (La Niña) years, and the remaining years are neutral years.

Precipitation and δ18Op data

The meteorological data of XRY (Yangshan station, 1957–2017 CE) is from the China Surface Climate Daily Dataset; data are averaged and summed to obtain monthly and annual data for analysis. HK δ18Op data during 1980–2017 comes from the global network for isotopes in precipitation (GNIP). The original data is converted into seasonal signal by 3-month-moving average to extract δ18Omin, and the annual δ18O is obtained by averaging original data.

Spatial precipitation and δ18Op data come from two datasets, CRU_TS_4.07 (1952-2017 CE) and IsoGSM2 (1979-2017 CE). IsoGSM2 is a water isotope-permitting general circulation model 103 that is considered to reproduce well annually variabilities of precipitation isotopic compositions across EASM regions 57, 104. Moisture tracing is facilitated through an Eulerian-based approach 105, 106. The detailed description of the model setup and moisture tagging design can be found in Kathayat et al. 107. The monthly spatial precipitation data (1952–2017 CE) are obtained from CRU_TS_4.07, which is a gridded time-series dataset covering all land areas (excluding Antarctica) at 0.5° resolution.

Data availability

XRY δ18O data are available in the East Asian Paleoenvironmental Science Database ( https://doi.org/10.12262/IEECAS.EAPSD2024004). Other data used in this study is publicly available: MEI and PDO are available at https://psl.noaa.gov/data/climateindices/list/. The GNIP data is provided by the International Atomic Energy Agency ( https://nucleus.iaea.org/wiser). The IsoGSM2 data can be obtained from http://isotope.iis.u-tokyo.ac.jp/~kei/tmp/isogsm2/. The WNA TC best track data is from the China Meteorological Administration tropical cyclone database at http://tcdata.typhoon.org.cn/zjljsjj.html. The CRU_TS_4.07 data is available at https://crudata.uea.ac.uk/cru/data/hrg/cru_ts_4.07/. Meteorological data of XRY is provided by China Meteorological Data Service Center at http://data.cma.cn/data/.

Code availability

The MATLAB code for CEEMDAN is available at http://perso.ens-lyon.fr/patrick.flandrin/emd.html.

1.
Jing, R. et al. Global population profile of tropical cyclone exposure from 2002 to 2019. Nature 626, 549–554 (2024).
2.
Burrows, K. et al. Health disparities among older adults following tropical cyclone exposure in Florida. Nat. Commun. 14, 2221 (2023).
3.
Ortiz, A. M. et al. Impact of tropical cyclones on food security, health and biodiversity. Bull. World Health Organ. 101, 152–154 (2023).
4.
Ghosh, S. & Mistri, B. Cyclone-induced coastal vulnerability, livelihood challenges and mitigation measures of Matla–Bidya inter-estuarine area, Indian Sundarban. Nat. Hazards 116, 3857–3878 (2023).
5.
Balaguru, K. et al. Increased U.S. coastal hurricane risk under climate change. Sci. Adv. 9, eadf0259 (2023).
6.
Wu, J., Gao, X., Zhu, Y., Shi, Y. & Giorgi, F. Projection of the future changes in tropical cyclone activity affecting East Asia over the Western North Pacific based on multi-RegCM4 simulations. Adv. Atmos. Sci. 39, 284–303 (2022).
7.
Holland, G. J. & Webster, P. J. Heightened tropical cyclone activity in the North Atlantic: natural variability or climate trend? Philos. T. R. Soc. A 365, 2695–2716 (2007).
8.
Landsea, C. W., Harper, B. A., Hoarau, K. & Knaff, J. A. Can we detect trends in extreme tropical cyclones? Science 313, 452–454 (2006).
9.
Lin, Y., Zhao, M. & Zhang, M. Tropical cyclone rainfall area controlled by relative sea surface temperature. Nat. Commun. 6, 6591 (2015).
10.
Trouet, V., Babst, F. & Meko, M. Recent enhanced high-summer North Atlantic Jet variability emerges from three-century context. Nat. Commun. 9, 180 (2018).
11.
Goslin, J. & Clemmensen, L. B. Proxy records of Holocene storm events in coastal barrier systems: storm-wave induced markers. Quat. Sci. Rev. 174, 80–119 (2017).
12.
Xian, H., Dong, X., Li, Y., Zhan, N. & Jeppesen, E. High-resolution reconstruction of typhoon events since similar to 1850 CE based on multi-proxy sediment records in a coastal lagoon, South China. Sci. Total Environ. 803, 150063 (2022).
13.
Oliva, F., Viau, A. E., Peros, M. C. & Bouchard, M. Paleotempestology database for the western North Atlantic basin. Holocene 28, 1664–1671 (2018).
14.
Soelen, E. E., Wagner-Cremer, F., Damste, J. S. S. & Reichart, G. J. Reconstructing tropical cyclone frequency using hydrogen isotope ratios of sedimentary n-alkanes in northern Queensland. Aust. Paleogeogr. Paleoclimatol. Paleoecol. 376, 66–72 (2013).
15.
Yang, Y. et al. Anomalously weak intensity of tropical cyclones striking eastern China over the past two millennia. npj Clim. Atmos. Sci. 7, 1–9 (2024).
16.
Wallace, D. J., Woodruff, J. D., Anderson, J. B. & Donnelly, J. P. Palaeohurricane reconstructions from sedimentary archives along the Gulf of Mexico, Caribbean Sea and western North Atlantic Ocean margins. in Sedimentary coastal zones from high to low latitudes: similarities and differences (eds Martini, I. P. & Wanless, H. R.) 388 481–501 (Geological Soc Publishing House, Bath, 2014).
17.
Zhao, N. et al. Daily growth rate variation in Tridacna shells as a record of tropical cyclones in the South China Sea: Palaeoecological implications. Paleogeogr. Paleoclimatol. Paleoecol. 615, 111444 (2023).
18.
Jiang, W. et al. Impacts of tropical cyclones and anthropogenic activities on marine vanadium: a unique perspective from high resolution Porites coral record. Sci. China Earth Sci. 65, 2285–2296 (2022).
19.
Kilbourne, K. H., Moyer, R. P., Quinn, T. M. & Grottoli, A. G. Testing coral-based tropical cyclone reconstructions: an example from Puerto Rico. Paleogeogr. Paleoclimatol. Paleoecol. 307, 90–97 (2011).
20.
Zhang, Q., Gu, X., Li, J., Shi, P. & Singh, V. P. The impact of tropical cyclones on extreme precipitation over coastal and inland areas of China and its association to ENSO. J. Clim. 31, 1865–1880 (2018).
21.
Czajkowski, J., Villarini, G., Michel-Kerjan, E. & Smith, J. A. Determining tropical cyclone inland flooding loss on a large scale through a new flood peak ratio-based methodology. Environ. Res. Lett. 8, 044056 (2013).
22.
Villarini, G., Goska, R., Smith, J. A. & Vecchi, G. A. North Atlantic tropical cyclones and U.S. flooding. B. Am. Meteorol. Soc. 95, 1381–1388 (2014).
23.
Chan, K. T. F., Zhang, K., Wu, Y. & Chan, J. C. L. Landfalling hurricane track modes and decay. Nature 606, E7–E11 (2022).
24.
Zhu, Y.-J. & Collins, J. M. Recent rebounding of the post-landfall hurricane wind decay period over the continental United States. Geophys. Res. Lett. 48, e2020GL092072 (2021).
25.
Song, J., Klotzbach, P. J., Zhao, H. & Duan, Y. Slowdown in the decay of western north pacific tropical cyclones making landfall on the Asian continent. Front. Earth Sci. 9, 749287 (2021).
26.
Lai, Y. et al. Slower-decaying tropical cyclones produce heavier precipitation over China. npj Clim. Atmos. Sci. 7, 1–13 (2024).
27.
Bregy, J. C. et al. US Gulf Coast tropical cyclone precipitation influenced by volcanism and the North Atlantic subtropical high. Commun. Earth Environ. 3, 1–11 (2022).
28.
Altman, J. et al. Poleward migration of the destructive effects of tropical cyclones during the 20th century. Proc. Natl Acad. Sci. USA 115, 11543–11548 (2018).
29.
Frappier, A. B. et al. Two millennia of tropical cyclone-induced mud layers in a northern Yucatán stalagmite: Multiple overlapping climatic hazards during the Maya Terminal Classic “megadroughts”. Geophys. Res. Lett. 41, 5148–5157 (2014).
30.
Frappier, A. B., Sahagian, D., Carpenter, S. J., González, L. A. & Frappier, B. R. Stalagmite stable isotope record of recent tropical cyclone events. Geology 35, 111 (2007).
31.
Cheng, H. et al. Chinese stalagmite paleoclimate researches: a review and perspective. Sci. China Earth Sci. 62, 1489–1513 (2019).
32.
Baldini, L. M. et al. Persistent northward North Atlantic tropical cyclone track migration over the past five centuries. Sci. Rep. 6, 37522 (2016).
33.
Frappier, A. B. Masking of interannual climate proxy signals by residual tropical cyclone rainwater: evidence and challenges for low-latitude speleothem paleoclimatology. Geochem. Geophys. Geosyst. 14, 3632–3647 (2013).
34.
Ansari, M. A., Noble, J., Deodhar, A. & Saravana Kumar, U. Atmospheric factors controlling the stable isotopes (δ18O and δ2H) of the Indian summer monsoon precipitation in a drying region of Eastern India. J. Hydrol. 584, 124636 (2020).
35.
Lawrence, J. R. & Gedzelman, S. D. Low stable isotope ratios of tropical cyclone rains. Geophys. Res. Lett. 23, 527–530 (1996).
36.
Xu, T. et al. Stable isotope ratios of typhoon rains in Fuzhou, Southeast China, during 2013-2017. J. Hydrol. 570, 445–453 (2019).
37.
Haig, J., Nott, J. & Reichart, G.-J. Australian tropical cyclone activity lower than at any time over the past 550–1,500 years. Nature 505, 667–671 (2014).
38.
Tang, H. et al. Mid-Holocene hydroclimatic change and hurricane activity in Central America recorded by an Isla de Mona Stalagmite. Mar. Geol. 471, 107289 (2024).
39.
Liu, T. S., Liu, S. J., Yang, X., Chen, J. & Zhang, C. Statistical analysis of the typhoon influencing Guangdong province during 1951-2015. Mar. Forecast. 35, 68–74 (2018).
40.
Sajjad, M. & Chan, J. C. L. Risk assessment for the sustainability of coastal communities: a preliminary study. Sci. Total Environ. 671, 339–350 (2019).
41.
Chen, F. et al. Typhoon control of precipitation dual isotopes in Southern China and its palaeoenvironmental implications. J. Geophys. Res. Atmos. 126, e2020JD034336 (2021).
42.
Yang, Y. et al. Typhoon rainfall impact on drip water δ18O in Xianyun cave, Southeast China. Hydrol. Process. 35 (2021).
43.
Zhang, H. et al. A 200-year annually laminated stalagmite record of precipitation seasonality in southeastern China and its linkages to ENSO and PDO. Sci. Rep. 8, 12344 (2018).
44.
Zhang, J. et al. Coupled effects of moisture transport pathway and convection on stable isotopes in precipitation across the East Asian monsoon region: implications for paleoclimate reconstruction. J. Clim. 34, 9811–9822 (2021).
45.
Hendy, C. H. The isotopic geochemistry of speleothems—I. The calculation of the effects of different modes of formation on the isotopic composition of speleothems and their applicability as palaeoclimatic indicators. Geochim. Cosmochim. Acta 35, 801–824 (1971).
46.
Dorale, J. A. & Liu, Z. Limitations of hendy test criteria in judging the paleoclimatic suitability of speleothems and the need for replication. J. Cave Karst Stud. 71, 73–80 (2009).
47.
Kim, S.-T. & O’Neil, J. R. Equilibrium and nonequilibrium oxygen isotope effects in synthetic carbonates. Geochim. Cosmochim. Acta 61, 3461–3475 (1997).
48.
Baker, A. J. et al. Seasonality of westerly moisture transport in the East Asian summer monsoon and its implications for interpreting precipitation δ18O. J. Geophys. Res. Atmos. 120, 5850–5862 (2015).
49.
Cai, Z., Tian, L. & Bowen, G. J. ENSO variability reflected in precipitation oxygen isotopes across the Asian Summer Monsoon region. Earth Planet. Sci. Lett. 475, 25–33 (2017).
50.
Li, J., Fan, L., Liang, M. & Xu, J. Water vapor transport differences of frontal and monsoon precipitation during South China pre-flood season under different PDO phases. Trans. Atmos. Sci. 45, 539–551 (2022).
51.
Qiu, W. et al. Oxygen isotope variability in precipitation, dripwater, and modern calcite responding to ENSO based on 11 years’ monitoring in Yuhua Cave, Southeast China. J. Hydrol. 624, 129937 (2023).
52.
Chakraborty, S. et al. Atmospheric controls on the precipitation isotopes over the Andaman Islands. Bay Bengal Sci. Rep. 6, 19555 (2016).
53.
Wei, Z. et al. Influences of large-scale convection and moisture source on monthly precipitation isotope ratios observed in Thailand, Southeast Asia. Earth Planet. Sci. Lett. 488, 181–192 (2018).
54.
Kyle, H. L., Weiss, M. & Ardanuy, P. Cloud, surface temperature, and outgoing longwave radiation for the period from 1979 to 1990. J. Clim. 8, 2644–2658 (1995).
55.
Ruan, J., Zhang, H., Cai, Z., Yang, X. & Yin, J. Regional controls on daily to interannual variations of precipitation isotope ratios in Southeast China: implications for paleomonsoon reconstruction. Earth Planet. Sci. Lett. 527, 115794 (2019).
56.
Lu, J. et al. A 120-year seasonally resolved speleothem record of precipitation seasonality from southeastern China. Quat. Sci. Rev. 264, 107023 (2021).
57.
Zhang, H. et al. Effect of precipitation seasonality on annual oxygen isotopic composition in the area of spring persistent rain in southeastern China and its paleoclimatic implication. Climate 16, 211–225 (2020).
58.
Zhao, J. et al. Reconstructing the western boundary variability of the Western Pacific Subtropical High over the past 200 years via Chinese cave oxygen isotope records. Clim. Dyn. 52, 3741–3757 (2019).
59.
Jackisch, D. et al. Precipitation stable isotopic signatures of tropical cyclones in Metropolitan Manila, Philippines, show significant negative isotopic excursions. Nat. Hazards Earth Syst. Sci. 22, 213–226 (2022).
60.
Munksgaard, N. C. et al. Stable isotope anatomy of tropical cyclone Ita, North-Eastern Australia, April 2014. Plos ONE 10, e0119728 (2015).
61.
Ohsawa, S. & Yusa, Y. Isotopic characteristics of typhonic rainwater: typhoons no. 13 (1993) and no. 6 (1996). Limnology 1, 143–149 (2000).
62.
Sun, C. et al. Isotopic variability in tropical cyclone precipitation is controlled by Rayleigh distillation and cloud microphysics. Commun. Earth Environ. 3, 50 (2022).
63.
Yang, Y., Xu, T., Luo, C., Liu, J. & Jiang, X. Analysis on the variation of typhoon precipitation δ18O during typical El Niño event: a case study of Typhoon Mangkhut (2018). J. Trop. Oceanogr. 39, 34–41 (2020).
64.
Konecky, B. L., Noone, D. C. & Cobb, K. M. The influence of competing hydroclimate processes on stable isotope ratios in tropical rainfall. Geophys. Res. Lett. 46, 1622–1633 (2019).
65.
Xu, T. et al. A tentative study of ‘Cloudy and Rainy Area Effect’ of the δ18O in the precipitation of typhoon ‘Dujuan’. J. Nat. Resour. 33, 2238–2248 (2018).
66.
Chang, C.-P., Zhang, Y. & Li, T. Interannual and interdecadal variations of the East Asian summer monsoon and tropical pacific SSTs. Part I: roles of the subtropical ridge. J. Clim. 13, 4310–4325 (2000).
67.
Li, C. Interaction between anomalous winter monsoon in East Asia and El Nino events. Adv. Atmos. Sci. 7, 36–46 (1990).
68.
Needham, H. F., Keim, B. D. & Sathiaraj, D. A review of tropical cyclone-generated storm surges: global data sources, observations, and impacts. Rev. Geophys. 53, 545–591 (2015).
69.
Jiang, H. & Zipser, E. J. Contribution of tropical cyclones to the global precipitation from eight seasons of TRMM Data: regional, seasonal, and interannual variations. J. Clim. 23, 1526–1543 (2010).
70.
Hu, F., Li, T., Liu, J., Bi, M. & Peng, M. Decrease of tropical cyclone genesis frequency in the western North Pacific since 1960s. Dyn. Atmos. Oceans 81, 42–50 (2018).
71.
Balaguru, K., Foltz, G. R., Leung, L. R. & Emanuel, K. A. Global warming-induced upper-ocean freshening and the intensification of super typhoons. Nat. Commun. 7, 1–8 (2016).
72.
Studholme, J., Fedorov, A. V., Gulev, S. K., Emanuel, K. & Hodges, K. Poleward expansion of tropical cyclone latitudes in warming climates. Nat. Geosci. 15, 14–28 (2022).
73.
Wadler, J. B. et al. A review of recent research progress on the effect of external influences on tropical cyclone intensity change. Trop. Cyclone Res. Rev. 12, 200–215 (2023).
74.
Murakami, H. Substantial global influence of anthropogenic aerosols on tropical cyclones over the past 40 years. Sci. Adv. 8, eabn9493 (2022).
75.
Wang, S., Murakami, H. & Cooke, W. F. Anthropogenic forcing changes coastal tropical cyclone frequency. npj Clim. Atmos. Sci. 6, 1–10 (2023).
76.
Zhao, J., Zhan, R., Wang, Y. & Xu, H. Contribution of the Interdecadal Pacific Oscillation to the recent abrupt decrease in tropical cyclone genesis frequency over the Western North Pacific since 1998. J. Clim. 31, 8211–8224 (2018).
77.
Liu, K. S. & Chan, J. C. L. Inactive period of Western North Pacific tropical cyclone activity in 1998–2011. J. Clim. 26, 2614–2630 (2013).
78.
Ren, F., Liang, J., Wu, G., Dong, W. & Yang, X. Reliability analysis of climate change of tropical cyclone activity over the Western North Pacific. J. Clim. 24, 5887–5898 (2011).
79.
Song, J.-J., Wang, Y. & Wu, L. Trend discrepancies among three best track data sets of western North Pacific tropical cyclones. J. Geophys. Res. Atmos. 115, D12128 (2010).
80.
Barcikowska, M., Feser, F., Zhang, W. & Mei, W. Changes in intense tropical cyclone activity for the western North Pacific during the last decades derived from a regional climate model simulation. Clim. Dyn. 49, 2931–2949 (2017).
81.
Hsu, P.-C. et al. Role of scale interactions in the abrupt change of tropical cyclone in autumn over the Western North Pacific. Clim. Dyn. 49, 3175–3192 (2017).
82.
Zhang, C. et al. Perspective on landfalling frequency and genesis location variations of Southern China typhoon during peak summer. Geophys. Res. Lett. 46, 6830–6838 (2019).
83.
Li, D., Jiang, Y., Zhang, L., Wang, H. & Li, X. Onset and retreat dates of the South China sea summer monsoon and their relationships with the monsoon intensity in the context of climate warming. J. Trop. Meteorol. 22, 362–373 (2016).
84.
Chan, J. C. L. & Zhou, W. PDO, ENSO and the early summer monsoon rainfall over south China. Geophys. Res. Lett. 32, L08810 (2005).
85.
Huang, R., Chen, J., Wang, L. & Lin, Z. Characteristics, processes, and causes of the spatio-temporal variabilities of the East Asian monsoon system. Adv. Atmos. Sci. 29, 910–942 (2012).
86.
Zhou, W. & Chan, J. C. L. ENSO and the South China Sea summer monsoon onset. Int. J. Climatol. 27, 157–167 (2007).
87.
Edwards, R. L., Chen, J. H. & Wasserburg, G. J. 238U234U230Th232Th systematics and the precise measurement of time over the past 500,000 years. Earth Planet. Sci. Lett. 81, 175–192 (1987).
88.
Cheng, H. et al. The half-lives of uranium-234 and thorium-230. Chem. Geol. 169, 17–33 (2000).
89.
Cheng, H. et al. Improvements in 230Th dating, 230Th and 234U half-life values, and U–Th isotopic measurements by multi-collector inductively coupled plasma mass spectrometry. Earth Planet. Sci. Lett. 371–372, 82–91 (2013).
90.
Breitenbach, S. F. M. et al. COnstructing proxy records from age models (COPRA). Climate 8, 1765–1779 (2012).
91.
Tan, L., Orland, I. & Cheng, H. Annually laminated speleothems in paleoclimate studies. Glob. Change Mag. 22, 22–23 (2014).
92.
Klotzbach, P. J. Trends in global tropical cyclone activity over the past twenty years (1986-2005). Geophys. Res. Lett. 33, L10805 (2006).
93.
Titus, M. L., Thompson, K. R., Oliver, E. C. J. & Klotzbach, P. J. Statistical Reconstruction of Seasonal Tropical Cyclone Variability in the North Atlantic Basin. J. Geophys. Res. Atmos. 126, e2020JD032669 (2021).
94.
Truchelut, R. E. et al. Earlier onset of North Atlantic hurricane season with warming oceans. Nat. Commun. 13, 4646 (2022).
95.
Lu, X., Yu, H., Yang, X. & Li, X. Estimating tropical cyclone size in the Northwestern Pacific from geostationary satellite infrared images. Remote Sens. 9, 728 (2017).
96.
Ying, M. et al. An overview of the China meteorological administration tropical cyclone database. J. Atmos. Ocean. Tech. 31, 287–301 (2014).
97.
Yin, Y. Analysis on the characteristics of typhoon affecting shenzhen city from 1961 to 2018. Adv. Meteorol. Sci. Tech. 12, 50–54 (2022).
98.
Colominas, M. A., Schlotthauer, G. & Torres, M. E. Improved complete ensemble EMD: a suitable tool for biomedical signal processing. Biomed. Signal Proces. 14, 19–29 (2014).
99.
Torres, M. E., Colominas, M. A., Schlotthauer, G. & Flandrin, P. A complete ensemble empirical mode decomposition with adaptive noise. in 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 4144–4147 (Prague, Czech Republic, 2011).
100.
Wolter, K. & Timlin, M. Monitoring ENSO in COADS with a seasonally adjusted principal component index. in Proceedings of the 17th Climate Diagnostics Workshop 52–57 (Norman, OK, 1993).
101.
Mazzarella, A., Giuliacci, A. & Scafetta, N. Quantifying the multivariate ENSO index (MEI) coupling to CO2 concentration and to the length of day variations. Theor. Appl. Climatol. 111, 601–607 (2013).
102.
Wolter, K. & Timlin, M. S. Measuring the strength of ENSO events: how does 1997/98 rank? Weather 53, 315–324 (1998).
103.
Yoshimura, K., Kanamitsu, M., Noone, D. & Oki, T. Historical isotope simulation using Reanalysis atmospheric data. J. Geophys. Res. 113, D19108 (2008).
104.
Chiang, J. C. H., Herman, M. J., Yoshimura, K. & Fung, I. Y. Enriched East Asian oxygen isotope of precipitation indicates reduced summer seasonality in regional climate and westerlies. Proc. Natl Acad. Sci. USA 117, 14745–14750 (2020).
105.
Tabor, C. R. et al. Interpreting precession-driven δ18O variability in the South Asian Monsoon Region. J. Geophys. Res. Atmos. 123, 5927–5946 (2018).
106.
Suwarman, R. et al. The variability of stable isotopes and water origin of precipitation over the maritime continent. Sola 9, 74–78 (2013).
107.
Kathayat, G. et al. Interannual oxygen isotope variability in Indian summer monsoon precipitation reflects changes in moisture sources. Commun. Earth Environ. 2, 96 (2021).
108.
Wang, S. et al. Spatial and seasonal isotope variability in precipitation across china: monthly isoscapes based on regionalized fuzzy clustering. J. Clim. 35, 3411–3425 (2022).
109.
Yang, X., Davis, M. E., Acharya, S. & Yao, T. Asian monsoon variations revealed from stable isotopes in precipitation. Clim. Dyn. 51, 2267–2283 (2018).

Acknowledgements

We thank Xiqian Wang, Dong Li for fieldwork assistance. We also acknowledge Prof. Wenxia Han for her help in δ18O analyses. This work was supported by the National Natural Science Foundation (42325705) and Chinese Academy of Sciences (xbzg-zdsys-202217 and 2020VCA0019).

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

About this article

Cite this article

Chen, L., Wang, T., Sinha, A. et al. A seasonally resolved stalagmite δ18O record indicates the regional activity of tropical cyclones in Southeast China. npj Clim Atmos Sci 7, 168 (2024). https://doi.org/10.1038/s41612-024-00715-0

-
Received:
-
Accepted:
-
Published:
-
DOI: https://doi.org/10.1038/s41612-024-00715-0