An Overview of the Holocene High Sea Level Around the South China Sea: Age, Height, and Mechanisms

Abstract

Understanding Holocene high sea levels in the South China Sea (SCS) is critical for understanding climate change and assessing future sea-level rise risks. We provide a comprehensive review of the Holocene highstand in the SCS, focusing on its age, height, and mechanisms. Records reveal a wide range for this highstand: ages span 3480–7500 cal yr BP, while elevations range from −7.40 to 7.53 m relative to the present. Positive elevations dominate (80.5% of records), with the most frequent range being 2–3 m. Regionally averaged formation times suggest a broadly synchronous mid-Holocene high-sea-level event across the SCS, potentially reflecting a global background. The observed variability is attributed to the interplay of multiple factors: global processes like glacial meltwater input and seawater thermal expansion, particularly during the Holocene warm period, and regional neotectonic movements (uplift/subsidence), which are the primary cause of spatial differences in reconstructed elevations. Significant debate persists regarding precise timing, height, and dominant mechanisms due to limitations in data coverage, dating precision, and challenges in quantifying tectonic influences. Future research priorities include obtaining high-resolution data from stable marine sediments, employing diverse dating techniques and modern crustal deformation monitoring, quantifying tectonic impacts, developing regional sea-level models, and enhancing international collaboration to refine understanding and improve predictions of future sea-level rise impacts.

1. Introduction

Rising global average sea levels driven by climatic warming are a pressing environmental concern which has captured the attention of governments, scientists, and the public worldwide [1,2,3,4,5]. The Intergovernmental Panel on Climate Change (IPCC) predicts that the global average sea level will rise by 0.52 to 0.98 m by 2100 [6], with the sea level of the South China Sea (SCS) expected to rise by ~30 cm and ~60 cm by 2030 and 2050, respectively [7]. These rises in sea level will not only submerge low-lying coastal areas, leading to significant shifts in coastlines and substantial changes in the ratio of land to sea area [8,9], but they will also exacerbate other environmental problems, including coastal erosion, seawater intrusion, storm surges, and soil salinization [10]. Approximately 37% of the global population resides within 100 km of a coastline [11]. These challenges threaten future socioeconomic development and even human survival, highlighting the importance of understanding past and future sea-level changes.

Within the context of Holocene relative sea-level (RSL) fluctuations, numerous studies suggested the occurrence of higher sea level relative to the present in the Asia and Oceania regions, with regional variations in the magnitude of these changes [12,13,14,15]. However, the specifics of when these peaks occurred, their heights, and the mechanisms behind them remain debated. Understanding these issues is crucial for predicting future sea-level changes and for informing coastal protection, utilization, and development policies globally.

2. Ages and Heights of the Holocene High Sea Level in the SCS

The SCS, a marginal sea situated at the transition zone between the Pacific Ocean and the Eurasian continent, is delimited by mainland China and Taiwan to the north, the Indochina and Malay Peninsulas and the island of Sumatra to the west, Borneo to the south, and Luzon and Palawan to the east [16]. Covering an area of ~3.5 million km², the SCS has an average water depth of ~1200 m and a maximum depth exceeding 5500 m [17]. Due to its position on the dynamic boundary between the Earth’s largest ocean and its largest continent, the SCS experiences the most vigorous material and energy exchanges between marine and terrestrial domains worldwide [18,19]. It has the most extensive tropical continental shelf, known as the Sunda Shelf, which makes the region highly susceptible to sea-level variations during glacial cycles. Even minor sea level movements can cause substantial changes in land–sea interactions, with significant regional environmental repercussions. Consequently, the SCS has been the focus of extensive research, especially concerning its high sea level during the Holocene epoch (Figure 1, Supplementary Table S1).

2.1. Ages and Heights of the Holocene High Sea Level in the Northern SCS

Previous studies suggest that the high sea level on Taiwan Island occurred at 5190 cal yr BP, with a height of 3.2 m (all elevation values are consistently referenced to present-day “modern mean sea level (MSL)” in this paper, and each reported elevation has been corrected for local vertical tectonic displacement) [20]. Others have argued that the high sea-level recorded by shells and coral reefs in Penghu, Taiwan, occurred at 4700 cal yr BP, with a maximum level of 2.4 m [21,69,70]. According to previous research, the RSL reached its maximum height of 2.3 m around 5000 cal yr BP in Taiwan [22]. The high sea level in the southern Fujian region occurred at 6350 cal yr BP, with a height of 4.5 m [23], while others concluded that it occurred between 7000 and 5700 cal yr BP, with a height of 3.0 m [46,71]. The high sea level determined by the dating of shell fragments and other materials in Eastern Guangdong was at 6300 cal yr BP, with a height of 2.0 m [23]. The Han River Delta experienced a highstand at 4790 cal yr BP, with a sea level only 0.7 m above the present-day level [72]. The highstand in Shenzhen, indicated by the elevation of rotted mangrove-related wood and oyster shells, ranged between −6.0 and −5.2 m and was established between 5440 and 5530 cal yr BP [26]. Hong Kong exhibited a highstand that was 4–5 m higher than today, formed at 5140 cal yr BP [27]; It was concluded that the mid-Holocene sea level was no higher than ~2 m above the MSL in Hong Kong [28]. Sediments on the Zhuhai shelf reveal a highstand of 3.8 m, at 5100 cal yr BP [73]. The Pearl River Delta also experienced a highstand at 5900 cal yr BP, with an elevation of 1.5 m [73].

Denglou Jiao in the Leizhou Peninsula was at an elevation of 3.7–3.9 m between 6550 and 5950 cal yr BP [29]. Investigations of the coral reef geomorphology of Dengjiao in the Leizhou Peninsula indicated that the exposed coral reefs in this tectonically stable area reliably indicate a sea level 2–3 m higher than today, between 7200 and 6700 yr BP (U-series dating) [30]. According to a study of seven coral reefs in the Leizhou Peninsula, it concluded that the Holocene high sea level in this region occurred between 7200 and 5000 yr BP (U-Th dating), and that it was at least 2.9–3.8 m higher than today [31]. Several researchers have proposed that the highest sea level on the Leizhou Peninsula occurred around 7000 years ago, based on U-series dating, with a highstand of 1.8 m [32]. It was reported that during 7050–6600 yr BP (U-series dating), the sea level was ~1.71–2.19 m higher than today at Leizhou Peninsula [33]. Beach rocks on Weizhou Island in Guangxi indicated a highstand of 6 m, which occurred at 6000 cal yr BP [34].

Overall, the high sea-level elevation records in the northern SCS include both positive and negative elevations, ranging from −7.4 to 6.0 m (average of ~0.9 m), with a very large range of elevation changes. The occurrence interval of the maximum Holocene sea level was between 3480 and 7200 yr BP (average of ~5690 cal yr BP).

2.2. Ages and Heights of the Holocene High Sea Level in Coastal Areas of Hainan Island

Extensive research has been conducted on the Holocene high sea levels of Hainan Island. Radiocarbon dates were obtained for two reefs: one at Luhuitou in Sanya with an elevation of 1.5 m, dated to 5180 ± 190 cal yr BP, and another on the west coast of Xiaodonghai in Sanya, with an elevation of 2.5 m, dated to 4800 ± 240 cal yr BP [36]. It was reported that a primary reef at Dongmaozhu in Sanya with an elevation of 0.8 m, and was dated to 6345 ± 100 cal yr BP based on a study of coral reefs at Luhuitou [36]. A comprehensive study of the Holocene sea level on Hainan Island and suggested that the high sea level occurred between 6000 and 5500 cal yr BP, with the level being 5.95–7.53 m higher than today [74]. This elevated estimate may be related to limitations in the precision of their methods. Specifically, their calculation of regional tectonic uplift or subsidence might not have been sufficiently accurate, resulting in significantly higher estimates compared to subsequent studies. According to research, the sea level was 4–5 m higher than today at Luhuitou in southern Hainan Island, between 6000 and 5000 cal yr BP [75]. A primary reef at Luhuitou was dated to 4800 ± 240 cal yr BP, with an elevation of 2.5 m [29]. Primary reefs at Luhuitou in Sanya, with elevations of 1.2 m and 3.0 m, were dated to 5025 ± 85 cal yr BP and 5000 ± 200 cal yr BP, respectively [37]. Coral reef studies indicated a high sea-level event between 7300 and 6000 cal yr BP in southern Hainan Island, with the maximum elevation at 0 m [38]. These records from the southern Sanya area of Hainan Island indicate a high sea level at ~5000 cal yr BP, with an elevation of ~2.5 m. According to a study of primary fossil corals and living micro-atolls in the eastern Qionghai area of Hainan Island, the sea level was at least 1.0 ± 0.08 m higher than today between 5500 and 5200 years ago, based on U/Th dating [41]. Research indicated that mid-Holocene sea level in Hainan Island fluctuated by up to 0.5 m, on the centennial scale, from 6143 to 4384 cal yr BP, based on coral microatoll fossils [76]. It was demonstrated that the sea-level rise in the Qiongzhou Strait, as inferred from the analysis of sediment cores and radiocarbon dating, peaked at 2–3 m above present levels during the period of approximately 9.0–6.0 ka BP [42]. Research on Hainan Island is predominantly focused on southern Hainan Island, in the Sanya area, with the high sea level occurring between 4800 and 7500 cal yr BP (average of ~5650 cal yr BP), over a broad time span, with an elevation ranging from 0 to 7.53 m (average of ~1.93 m).

2.3. Ages and Heights of the Holocene High Sea Level in Coastal Areas of Southeast Asia

The sea level peaked at Duy Tien in the Red River delta at ~6500 cal yr BP, registering a slightly higher stand than today, due to hydro-isostatic adjustments [77,78]. In the Red River Delta at Gia Lu, the high sea level was dated to 5730 cal yr BP, with an elevation of −1.6 m [44]. During the Holocene, a sea-level highstand of 2–3 m above the present level occurred between 6000 and 4000 cal yr BP, which facilitated extensive mangrove colonization in the Red River delta and the creation of marine notches in the regions of Ha Long Bay and Ninh Binh [79].

In the middle and lower plains of the Chao Phraya Delta, the high sea level occurred at 6300 cal yr BP, with an elevation of −6.0 m [48]. In the vicinity of the Mekong River Delta, at Con Dao and Bai Kan islands, the high sea level was dated to between 5810 and 6200 cal yr BP, also with an elevation of 5.0 m [43]. Based on detailed mapping and sedimentological studies, it was reported that in the Mekong River Delta the maximum post-glacial transgression was at ~6000–5000 BP, with the relative sea level reaching ~4.5 m above the modern level [45]. In Ho Chi Minh City, part of the Mekong River Delta, the high sea level occurred at 6530 cal yr BP, with an elevation of −0.4 m, and at Ta Me it occurred at 5500 cal yr BP, with an elevation of −1.0 m [46]. Within the Chao Phraya Delta, the high sea level occurred between 5600 and 5700 cal yr BP, with elevations ranging from 2.0 to 3.5 m [46]. The Holocene high sea-level stand at Cape Ga Na, Vietnam, occurred at ~4500 cal yr BP, with an elevation of 5.0 m [46]. In Southeast Vietnam, it was reported a mid-Holocene sea-level highstand between 1.4 and 1.6 m above the present level during 6700–5000 cal yr BP, based on an analysis of beach rock and beach-ridge deposits [47]. Collectively, the high sea levels of the Indochina Peninsula occurred mainly between 4000 and 6700 cal yr BP, with elevations ranging from −6.0 to 5.0 m.

In the Malay Peninsula, the Holocene high sea-level stand at Linggi Island and Perlis occurred between 5090 and 5200 cal yr BP, with elevations of 2.4 m and 3.0 m, respectively [52]. It was reported that the sea level (low spring tide) was ~1 m higher than today in Phuket, South Thailand, ~6000 years ago [49]. In the Kedah region, the high sea-level stand occurred at 5570 cal yr BP, with elevations primarily between 2.0 and 3.0 m [80], with Bujok Ran in the southern part of the peninsula recording an elevation of 3.0 m, and a negative elevation of −3.7 m. According to previous research, the relative high sea level at ~7000 cal yr BP in northeastern Peninsular Malaysia was between 1.4 and 3.0 m above the present level [55]. They also suggested that the regional relative sea-level data indicate a maximum Holocene highstand of up to 5 m between 6000 and 4000 cal yr BP. In the eastern region, the high sea level occurred at 4500 cal yr BP, with an elevation of 2.0 m [81].

In the Straits of Malacca, the elevation was 0.7 m [51]. The timing of the rise in high sea-level rise in Singapore is poorly constrained, estimated to be between ~7400 and ~7200 cal yr BP, with a magnitude of 3 to 5 m [57]. Previous research concluded that the RSL in Singapore reached its maximum elevation of 3.3  ±  0.3 m at 5800 cal yr BP [14,58].

Utilizing 15 new sea-level index points from Kuantan, in conjunction with recently published data from Terengganu, the RSL history of the eastern coast of the Malay Peninsula during the middle to late Holocene was reconstructed [81]. Their record indicates that the RSL was ~1.5 m above the MSL at ~6500 cal yr BP, after which it increased to ~2.0 m MSL by 4500 cal yr BP. Overall, the high sea level of the Malay Peninsula occurred between 4500 and 6000 cal yr BP, with elevations ranging from −3.7 to 3.0 m [82].

Much Holocene high sea-level research has also been conducted in the southern SCS region. For instance, on Java Island, the maximum recorded sea level in South Anyer County was 3.5 m. On Sumba Island, a high sea-level event occurred at 6300 cal yr BP, with an elevation of 3.0 m [60]. The elevated corals at Teluk Awur in the Sunda Shelf indicate that the early to mid-Holocene sea levels were ~0.5–1.5 m higher than today, at ~7000 cal yr BP [62]. In Makassar, Sulawesi, the Holocene high sea-level event occurred at 4902 cal yr BP, with an elevation of 6.5 m [52]. Fossil microatolls from Pulau Panambungan in the Strait of Makassar suggest a 0.5 m RSL highstand at ~5600 cal yr BP [59]. Similar elevations of 6.5 m were observed in Pangkalinan, while in the Brunei Limbang Valley, the elevation was 1.8 m [83]. On Minggiran Island, the high sea-level event occurred at 6260 cal yr BP, with an elevation of −0.7 m; on Tambelan Island, it was 0.4 m [52]; and on northwestern Belitung Island the Holocene high sea-level event occurred at ~6800 cal yr BP, with an elevation of 1.9 m [8]. Chronological and elevational analysis of microatolls on Natuna Island in Indonesia revealed a period of relatively high RSL stability between 6400 and 1400 cal yr BP, averaging 0.2–0.7 ± 0.4 m above the modern level, prior to a subsequent decline [63]. Overall, the high sea levels in the Sunda Shelf region ranged from −3.7 to 6.5 m, occurring between 4900 and 7400 cal yr BP.

In the Philippines, the Holocene high sea level in the Palawan region occurred between 5690 and 6700 cal yr BP, with elevations between 0.3 and 1.5 m. In the Luzon Island region, the high sea level occurred between 5073 and 7269 cal yr BP, with elevations between 0.9 and 2.9 m. Similarly, in the Samar Island region, the high sea level occurred between 4180 and 7255 cal yr BP, with elevations between 0.9 and 2.9 m [64]. In the Panglao Island region, the high sea level occurred during the mid-Holocene, 0.3–0.6 m higher than today [66]. The reef at Currimao, in northwestern Luzon Island, reached its high level of ~0.5 m above MSL at ~6860 cal yr BP [65]. Collectively, the Holocene high sea levels in the Philippines are all positive, ranging from 0.3 to 2.9 m, and they developed between 3570 and 7250 cal yr BP.

2.4. Ages and Heights of the Holocene High Sea Level in the SCS Basin Area

A systematic investigation of the elevations of coral reefs in the Nansha and Xisha Islands was conducted [84]. The reef tops in the Nansha Islands are generally dated at 4930 ± 160 cal yr BP, standing 0.3–0.4 m above the mean low water level of spring tides, with a few reaching 0.7–0.8 m. On Yongxing Island, one of the Xisha Islands, the reef top developed at 4000 cal yr BP, 0.6–0.8 m above the mean low water level of spring tides. The actual growth limit of coral is at a depth of more than 1 m below the low tide level during high tide, and the exposed top of the reef has been reduced in height by at least 20–30 cm due to long-term erosion. Therefore, the actual height of the reef top should be increased by at least 1.2 m to obtain the high sea surface at that time, which is about 1–2 m higher than today in the Xisha Islands. Several researchers have suggested that the high sea level represented by primary and secondary reef samples from the Xisha Islands was 0–1 m, between 4810 and 6400 cal yr BP [52,54]. Coral reef data from Chenhang Island indicate a sea-level highstand ~2 m above the modern MSL in the Xisha Islands during 6000–3900 cal yr BP [68]. Overall, the Holocene high sea level in the SCS Basin developed mainly between 4810 and 6400 cal yr BP (average of ~5410 cal yr BP), with significant elevation differences, ranging from 0 to 2.0 m (average of ~1.0 m).

2.5. Summary

The Holocene high sea-level records from the areas around the SCS show an age range from a maximum of 7500 cal yr BP to a minimum of 3480 cal yr BP, with an average of 5660 cal yr BP. The elevations range from a maximum of 7.53 m to a minimum of −7.4 m, with an average of 1.55 m. There are 74 instances (80.5%) of positive elevations, 6 instances (6.5%) of 0 m elevation, and 12 instances (13.0%) of negative elevations.

The most frequent (mode) of high sea-level elevation is between 2 and 3 m, with 20 records, accounting for 21.7% (Figure 2 and Figure 3).

As can be seen from the above summary, there are significant discrepancies in the reconstruction results of high sea levels in the SCS during the Holocene. The reason for this may be related to the methods used for sea level reconstruction, among which the most critical ones include the relationship between sea level marker elevation and sea level, as well as the adopted dating method and its correction. Despite the differences in the time and height of sea level in different regions, the average height (1.93 m) and average formation time (5650 cal yr BP) of the Holocene high sea level in Hainan Island are basically consistent with the average height (1.95 m) and average formation time (5660 cal yr BP) of the Holocene high sea level in Southeast Asia (Figure 4). This indicates that the Holocene high sea level in Hainan Island was basically synchronous with that in Southeast Asia. The average formation times of the Holocene high sea level in four different regions (northern SCS, Hainan Island, Southeast Asia, and SCS Basin) are 5690 cal yr BP, 5650 cal yr BP, 5660 cal yr BP, and 5410 cal yr BP, respectively, all of which are relatively close (Figure 4), suggesting that the Holocene high sea level in the SCS may have formed during the mid-Holocene and it is likely to have a global background.

Various techniques have been used to date the maximum Holocene sea level, including ¹⁴C dating, Thermal Ionization Mass Spectrometry (TIMS) U/Th (uranium-series) dating, and Optically Stimulated Luminescence (OSL) dating. The precision varies among the different dating methods [85]. Although ¹⁴C dating has high precision, most of the dated materials are of marine origin, making them subject to the marine carbon reservoir and old carbon effects. Moreover, the dated materials are diverse, including carbonates such as shells, oyster shells, beach rock, and coral reefs [86], and organic carbon materials have also been used, including mangrove wood, mud, and peat (Supplementary Table S1) [24,48,50]. This inconsistency in dating methods and materials inevitably leads to significant differences in age results. Additionally, some publications report age ranges, while others do not, making precise comparisons difficult.

It is important to note that the current elevations of ancient sea-level remnants do not directly correspond to the heights of the ancient sea level. In terms of the height of the maximum Holocene sea level, some researchers have considered factors such as tectonic uplift and subsidence, while others use the existing elevation of paleo-sea-level relics. Due to the influence of land surface uplift and subsidence, the existing elevation of paleo-sea-level relics may not reflect their true height. Moreover, it is very difficult to accurately estimate tectonic uplift and subsidence activity within the Holocene, and there is a lack of high-precision crustal vertical deformation data. Hence, it is difficult to eliminate the influence of tectonic uplift and subsidence in obtaining precise paleo-sea-level elevations.

3. Causal Mechanism of the Holocene High Sea Level in the SCS

Research on the mechanisms of Holocene high sea level has revealed various causal mechanisms across different scales of sea-level change [87]. Current understanding identifies three primary direct influencing factors on Holocene sea-level fluctuations: firstly, changes in the mass of seawater, such as those caused by glacier meltwater input; secondly, variations in the volume of seawater, primarily due to thermal expansion and contraction in response to temperature fluctuations [88]; and thirdly, crustal movements, including land subsidence, leading to relative sea-level changes [89,90,91].

3.1. Glacier Melting Effects on the Holocene Relative High Sea Level in the SCS

The influence of glacier melting on the RSL highstand during the Holocene in the SCS is complex, modulated by various geophysical processes. Since the Last Glacial Maximum, the melting of land-based ice has contributed to global mean sea-level rise through the addition of meltwater to the oceans [92]. Spatial patterns of RSL change—defined as sea level at a specific location relative to the present day—exhibit significant variability. This heterogeneity primarily arises from glacial isostatic adjustment (GIA), the delayed viscoelastic response of the solid Earth, its gravity field, and Earth’s rotation to the redistribution of ice and water loads [14,93,94].

During the early Holocene (until ~7000 yr BP), rapid meltwater influx from collapsing ice sheets (e.g., Laurentide, Antarctic) during deglaciation significantly elevated global eustatic sea level. Along the Fujian coast of South China, this meltwater pulse was the primary driver of RSL rise until approximately 7000 yr BP [24].

In the mid-to-late Holocene (after ~7000 yr BP), as ice unloading diminished, the GIA response became increasingly dominant. For instance, along the Fujian coast, a shift from meltwater-driven RSL rise to GIA-controlled subsidence occurred after ~7000 yr BP [88]. A study investigated the mid-Holocene sea-level highstand in Singapore and Southeast Asia, assessing its sensitivity to Earth and ice model parameters within GIA models [95]. Their findings demonstrate that variations in the Earth model primarily influence the highstand’s magnitude, whereas changes in the ice model affect both its timing and magnitude. They conclude that the mid-Holocene highstand is sensitive to upper and lower mantle viscosities and the ice-equivalent sea-level history, providing insights for refining GIA models. Furthermore, a detailed analysis of mid-Holocene RSL fluctuations in Southeast Asia was presented, using high-resolution proxy records from coral microatolls on Belitung Island. This analysis revealed significant centennial-scale RSL changes of up to 0.6 m between ~6850 and ~6500 yr BP [95]. After ~7000 yr BP, GIA emerged as the dominant mechanism controlling RSL change in the region [8]. GIA models also have limitations; for example, the ANICE-SELEN model—validated against Makassar Strait RSL data—predicts lower highstands than other models, underscoring the critical need for precise Earth and ice model parameters to accurately forecast modern vertical land motion [96].

Glacier melting drove early Holocene RSL rise via meltwater pulses (~7 ka BP), but post−7 ka BP, GIA became the dominant control on RSL variability in the SCS, with model accuracy constrained by parameter sensitivity and inherent limitations.

3.2. Temperature Effects on the Holocene Relative High Sea Level in the SCS

Temperature is crucial for determining Holocene sea level in the SCS. One of the primary mechanisms by which temperature affects sea level is via the thermal expansion and contraction of seawater. As global temperatures rise, seawater expands, leading to an increase in sea level; conversely, during cooler periods, seawater contracts, resulting in lower sea level. This thermosteric sea-level change is a critical component of the RSL changes observed in the SCS [88]. The high sea level in the SCS generally occurred between 7500 and 3480 cal yr BP (Supplementary Table S1); this interval is coeval with the East Asian Holocene Climatic Optimum (8000–4000 cal BP) defined by multi-proxy reconstructions [97,98,99,100,101,102]. The mean annual temperature in the Holocene Climatic Optimum in South China and adjacent subtropical East Asia was about 1–2 °C higher than that of the present day [103]. Rising temperatures cause glaciers to melt, which in turn causes sea levels to rise. This relationship implies that temperature is a fundamental controller of sea level in the SCS. As discussed above, temperature can influence the influx of meltwater from retreating glaciers and ice sheets which can contribute to sea-level rise [92]. While the SCS is far from the major ice sheets, global sea-level changes can still impact the region via the redistribution of water in the world’s oceans. The SCS is also influenced by monsoonal climate patterns, which cause seasonal variations in precipitation and temperature. These variations can affect local sea level via changes in the amount of freshwater influx from rivers and the intensity of storms [7]. Various paleoclimate proxy records from the SCS, including from coral cores and sediment cores, have provided insights into past sea level and the associated climatic conditions [8,15]. However, previous studies have yielded a relatively broad age range for the formation of the Holocene high sea level in the SCS, making it difficult to determine the phase relationship between the maximum Holocene sea level and the maximum global Holocene temperature. Specifically, did the maxima in sea level and global temperature occur synchronously, or was there a lead or lag effect? Furthermore, the significant maximum sea-level variations across different regions (Figure 1, Supplementary Table S1) may reflect diverse regional tectonic settings.

3.3. Neotectonic Effects on the Holocene Relative High Sea Level in the SCS

In the coastal region of Fujian and within the Taiwan Strait, minor seismic and tectonic uplifts have been detected. The Han River Delta was a sedimentary depocenter during the late Quaternary. This area has undergone significant subsidence, attributed to factors such as tectonic movements along fault lines [104,105]. Eastern Guangdong, with its relatively narrow continental shelf, is minimally influenced by tectonic activity and is considered to be in a state of relative stability, with only minor uplift occurring. The Pearl River Delta has also undergone minor subsidence, primarily due to fault activity [106]. In contrast, Western Guangdong, located at a greater distance from the Philippine plate boundary, exhibits greater tectonic stability [20,70]. Previous research concluded the occurrence of tectonic stability, with minimal vertical movements, since the mid- Holocene in the Leizhou Peninsula [33]. Most of these findings are based on research conducted mainly between the early 1980s and mid-1990s. While these historical data provide valuable insights into the geological history of these regions, it is essential to remember that geological processes can evolve over time due to factors such as climate change and tectonic movements [33]. Therefore, to comprehensively understand the current geological conditions in these areas, it is important to reference recent studies and use up-to-date data. It should also be noted that while the available information provides a detailed overview of the tectonic activity in these regions, further analysis may be needed to fully understand the complex interactions between different geological processes and their impacts on the landscape and environment.

In the northern region of Hainan Island, previous researches concluded that Holocene tectonic activity is relatively limited, with no significant vertical crustal movements [39,70]. However, other researchers have proposed that neotectonic activity in the northern part of Hainan Island has been more dynamic due to the expansion of the SCS. This region experienced multiple phases of intermittent volcanic eruptions during the Holocene. Notably, the $\mathrm{M} 7{\displaystyle \frac{1}{2}}$ magnitude earthquake in Qiongshan, Hainan, in AD 1605, was the most devastating earthquake in the northern SCS’s seismic belt, leading to the subsidence of over 100 km² of land and highlighting the occurrence of frequent active tectonics in this area [107,108]. Furthermore, the emergence of dead coral formations and the variations in their elevations can be attributed to tectonic uplift, potentially linked to volcanic activity along the northwestern coast of Hainan Island. When constructing high-resolution sea-level histories, it is crucial to consider the impact of vertical tectonics, even in regions like the SCS coast, which are generally regarded as tectonically stable [109]. Moreover, there is often considerable debate about the nature of tectonic movements within the same geographical area.

The Red River delta was tectonically stable during the Late Pleistocene and Holocene, despite the existence of the Red River fault system; however, tide gauge data indicate a relative sea-level rise of 2.24 mm/a over the past four decades [110,111]. The geological structure of the Gulf of Thailand is characterized by a horst-and-graben configuration, with active faults during the Holocene that significantly influence the current drainage patterns and topography of this region [48]. The Malay Peninsula, known as the Mesozoic Sundaland Core, is a region of relative tectonic quiescence, with vertical crustal movements that are two to three orders of magnitude lower than those in more tectonically active areas [53,112]. Vertical crustal movement likely varies from 0.3 to 0.5 mm/a along the East Coast and reaches 0.7 mm/a along the West Coast of Peninsular Malaysia [113].

The Sunda Shelf, encompassing the shallow marine areas between the Indochina Peninsula, Malay Peninsula, Sumatra, Java, and Borneo, constitutes the southernmost seabed of the SCS. This extensive platform forms part of the coastal regions of Cambodia, Thailand, Malaysia, Singapore, Borneo, and parts of the coasts of Indonesia and Vietnam. The Sunda Shelf, featuring a gently undulating surface topography, experienced stable tectonic conditions throughout the Quaternary [114], with minimal internal deformation. The platform is predominantly defined by its anomalous stability, and during lower sea level, it supported well-developed fluvial systems [115]. Currently, the Sunda block exhibits an extremely low rate of vertical uplift of <0.1 mm/a [53,116,117]. Based on the above, it can be seen that even for the same region, there is debate among different researchers about its tectonic uplift/subsidence, and this inconsistent understanding can affect the determination of high sea level height. The Sunda Shelf, at tropical low latitudes, is largely unaffected by GIA. During the mid-Holocene, the rate of glacial melting subsided, enhancing the prominence of water equilibrium effects, which potentially contribute to the RSL fluctuations observed in this region [88].

The Indonesian archipelago in Southeast Asia has experienced significant tectonic activity throughout the Holocene. This region is a collision zone for the Eurasian, Indian, Australian, Philippine, and Pacific plates, encompassing the active subduction zone and island arc associated with the Java Trench, which extends across a substantial portion of southern Indonesia [53].

The Philippine Islands are a geologically complex region, formed by the amalgamation of volcanic arcs, marginal basins, ophiolites, and continental fragments due to subduction, collisions, and strike–slip faulting [118]. The oblique convergence of the Philippine Plate with the Eurasian Plate is accommodated by two major collision zones—the Manila Trench on the west and the Philippine Trench in the east—as well as the strike–slip Philippine Fault [119] that spans over 1200 km from southern Mindanao to northern Luzon [120]. In Ilocos, the variability of paleo-sea-level data is attributed to segmentation and variable uplift across several northwest- and northeast-trending sinistral strike–slip faults cutting through the coast. These faults are active and may cause a clockwise rotation of the fault-bounded blocks with a southward tilt. In Palawan, no active faults have been identified, but abrupt changes in reconstructed paleo-sea levels suggest possible faulting. Uplift or movement along a fault likely caused the north-northeast and south-southwest tilting of northern and southern Palawan. In Samar, local faults explain discrepancies in paleo-sea levels along adjacent sites. Active faults in this island are likely controlled by the adjacent major structures, such as the Philippine Fault to the west and the Philippine Trench to the east. Uplift due to subduction along the Philippine Trench may explain the apparent westward tilt of Samar. Variation in regional tectonics and the response of the crust to hydro-isostasy explain the differences in paleo-sea levels within the Philippines: relatively high paleo-sea levels in Ilocos, intermediate levels in Samar, and the lowest levels in Palawan. Significant seismicity in northwest Luzon and Samar may contribute to uplift, while Palawan exhibits relative stability compared to Ilocos and Samar [64,121].

The SCS Basin was a subsiding area during the Cenozoic, yet the characteristics of neotectonic movements vary across different regions, leading to differential impacts on relative sea level. The modern tectonic movement of the Dongsha Uplift appears to be relatively stable. In the late Quaternary, the crust of the Nansha Islands appears to have risen, forming coral islands with an elevation of ~10 m, and the modern tectonic movement is likely of a minor ascending nature [122]. Based on a summary of coral reefs information, a subsidence trend in the crust of the Xisha Islands and other atolls was observed, with rates ranging from 0.07 to 0.10 mm/a [123]. A detailed analysis of single-channel seismic profiles from the Xisha Islands region concluded that the tectonic environment has become more stable after the Pleistocene, with only a few faults remaining active at present [124]. Similarly, the tectonics of the Chenhang Island area of the Xisha Islands were relatively stable during the Holocene [125].

In summary, the Holocene high sea levels in the SCS were influenced by a combination of factors including glacier melting (GIA), thermal expansion, and regional tectonic activity with significant variations observed across different coastal regions due to diverse geological settings and neotectonic processes.

3.4. Comparison of Holocene High Sea Levels in the SCS with Global Records

During the Mid-Holocene (approximately 8000–4000 cal yr BP), phase-specific high sea levels occurred in numerous locations worldwide, such as Western India [126], the west coast of South Africa [127], Brazil in South America [128], the Australian coast [129,130], and the Marshall Islands in the Central Pacific [131]. Geological evidence from these regions indicates that sea levels were several meters higher than the present at that time. This phenomenon was particularly prevalent in tropical and subtropical regions. However, it is noteworthy that mid-latitude regions of the Northern Hemisphere exhibit different characteristics. For example, sedimentary records from the Bay of Biscay [132], Florida [133] and generally along European [134] and U.S. Atlantic coasts [135,136] show that current sea levels are the highest of the past 10,000 years. Consequently, research on global Holocene sea-level evolution reveals significant regional heterogeneity.

The IPCC AR6 report provides the latest projections and assessments of global and regional sea-level change [137]. It emphasizes that while sea-level rise is global, its regional manifestation is influenced by multiple factors, including thermal expansion, ice sheet and glacier melt, changes in land water storage, and vertical crustal movement [138].

Globally, the South China Sea’s Holocene high sea levels reflect both global sea-level rise and distinct regional characteristics. Influenced by local tectonic activity, hydrological cycles, and ocean dynamics, they differ in timing and magnitude from global means and other regions (e.g., Atlantic). The IPCC AR6 report highlights this regional complexity, consistent with South China Sea observations. Thus, the SCS high stands resulted from regional responses superimposed on the global rise.

4. Problems and Future Prospects

Extensive research on Holocene high sea levels around the SCS has been conducted in previous studies, providing valuable data and establishing a solid foundation for future research. However, several issues persist.

(1) Uneven record distribution: The distribution of geological records in the SCS region is uneven. In some areas (like the Nansha Islands, Northern Hainan Island, etc.), data is scarce, which results in a low spatiotemporal resolution for reconstructing sea-level changes. Research conclusions from different regions are contradictory, leading to debate on the timing and height of high sea levels. (2) Limitations of dating precision and methods: Relying on ¹⁴C dating may lead to age errors due to the carbon reservoir effect (such as regional differences in the marine carbon reservoir). Biostatigraphic markers (such as the upper limit of coral growth) may be affected by later geological activities or erosion and could be misinterpreted as paleo-sea-level indicators. (3) Interference from tectonic activity and local subsidence: Plate tectonics around the SCS (such as subduction zones and volcanic activities) and regional crustal deformations (such as delta subsidence) may mask the true paleo-sea-level signal. The relationship between the vertical growth rate of coral reefs and the rate of tectonic uplift/subsidence has not been fully clarified. (4) Uncertainty in driving mechanisms: The connection between the causes of the Holocene high-sea-level stand (such as the rate of glacial melting and GIA) and regional climate responses (such as monsoon changes and El Niño-Southern Oscillation (ENSO) activity) still needs further validation. There is an issue of the applicability of global sea level change models (such as the contribution of ice-sheet melting) in the SCS region.

To address these issues, we suggest that future research focus on the following: (1) High-resolution data acquisition and sharing: conduct systematic geological drilling in less-disputed areas of the SCS to obtain continuous sedimentary sequences and coral reef samples. (2) Technological innovation and cross-validation of multiple indicators: Combine uranium-series dating (U-Th), OSL, and other supplementary methods with ¹⁴C dating to reduce chronological errors. Critically evaluate data precision, including reservoir corrections for marine carbonates and tectonic influences. Combine multiple lines of evidence, such as microfossil (foraminifera, diatoms) and geochemical indicators, to enhance the reliability of paleo-sea-level reconstruction. (3) Tectonic-sea-level coupling studies: Use GPS and InSAR technologies to monitor modern crustal deformation, and combine paleoseismic records to quantify the impact of tectonic activity on paleo-sea-level indicators. Develop regional GIA models to distinguish between glacial and tectonic related sea-level change signals. (4) Dynamic simulation of climate-sea-level interactions: Develop high resolution regional climate-sea-level coupled models that integrate regional climate factors such as monsoons, ocean currents, and ENSO. Compare high-sea-level records of the SCS with those of other marginal seas around the world to reveal regional-specific mechanisms. (5) Enhancing international cooperation: Establish an SCS paleo-sea-level database and promote transnational data sharing (such as cooperation with countries around the SCS).