Introduction

Over the past two centuries, human activities, such as industrialization, urbanization, globalization, and population growth, have irrefutably altered the Earth’s geological history, leading to substantial chemical disruptions and the onset of global climate change, including global warming1. By reconstructing element concentrations and accumulation rates in natural archives, such as ice cores and sediment records, we can gain a unique insight into how humans have responded to these challenges and predict the potential influence of these changes on the future2.

The Red Sea is a subtropical oligotrophic sea. Its nutrient sources primarily depend on vertical mixing in the northern region and wind-induced horizontal intrusion of nutrient-rich waters from the Indian Ocean in the southern region3. However, due to its semi-enclosed nature, infrequent precipitation, lack of riverine inputs, and hot climate, the Red Sea is considered one of the warmest sea basins globally, making it highly susceptible to warming4,5. Additionally, the Red Sea serves as a natural analog for potential future scenarios of overheated seas in the context of global warming.

Recent studies have revealed that the maximum surface temperatures of the Red Sea have been increasing at an alarming rate of 0.17 ± 0.07 °C per decade, surpassing the global ocean warming rate of 0.11 °C per decade4. This rapid warming has already led to anomalous heat waves and severe coral bleaching events6,7. Moreover, the warming trend is anticipated to exacerbate the severity of ocean stratification, impeding vertical exchanges between the surface and deep water layers, and curtailing nutrient availability for primary production, ultimately leading to food shortages for fish and other marine organisms8,9,10,11. It can also influence the interchange of water with the Indian Ocean12,13.

In addition to warming, the Red Sea is susceptible to various anthropogenic disturbances, including trace metal pollution14,15,16. The opening of the Suez Canal in 1869 coincided with the beginning of the Second Industrial Revolution between 1870 and 1914. This event, along with the discovery of the first Saudi Arabian oil field in 1938, contributed to a surge in commercial and oil shipping activities17, extensive development of oil-related industries18, and rapid urbanization and population growth in coastal cities along the Red Sea. These human activities may have released considerable amounts of trace metals into the Red Sea.

Sediment cores serve as the ultimate repository of trace metals in aquatic environments, offering a historical record of environmental changes and human influence over extended periods19. In the Red Sea, where riverine inputs are absent, and precipitation/runoff is negligible, sediment cores provide a clear and decipherable historical record. However, despite this potential, few studies have reported element concentrations in Saudi Arabian Red Sea sediment cores20,21,22,23, with most studies focused on coastal sediments and lacking supplemental chronological analysis. Thus, the spatial and temporal variability of element concentrations in the open Red Sea remains largely unknown.

The study examines the spatial and temporal variability of the 15 most representative elements in sediment cores collected from the North and South open Red Sea over the past centuries. The study determines whether the observed variations are due to natural forces or anthropogenic activities. The primary aim of this research is to establish a robust baseline for evaluating the long-term effects of human activities on marine environments, locally and globally. Additionally, the study aims to provide policymakers with targeted mitigation strategies to preserve and protect the fragile marine ecosystem in the Red Sea.

Results and Discussion

Declining nutrient input from the Indian Ocean to the Red Sea

The chronological analyses indicate that the dates of sediment cores range from the 1590 s to the 2010s. The spatial and temporal changes in 15 element concentrations in the Red Sea sediment cores are depicted in Fig. 1. The measured element concentrations are ranked in descending order by their mean values as follows: calcium (Ca) > silica (Si) > total organic carbon (TOC) > sodium (Na) > magnesium (Mg) > iron (Fe) > sulfur (S) > total nitrogen (TN) > phosphorus (P) > vanadium (V) > zinc (Zn) > nickel (Ni) > chromium (Cr) > copper (Cu) > cadmium (Cd). Further statistical analysis revealed that the South Red Sea sediment cores exhibited significantly higher element concentrations (excluding Ca and Mg; Table 1) but remarkably lower element accumulation rates (EARs, excluding TOC and TN, Fig. 2, Supplementary Table 1) compared to those of the North Red Sea (p < 0.01).

Fig. 1: Scatter plots of element concentrations in the Red Sea sediment cores.
figure 1

a–o Total organic carbon (TOC), total nitrogen (TN), Na, Mg, Si, P, S, Ca, V, Cr, Fe, Ni, Cu, Zn, and Cd concentrations (mg kg−1, unless noted as %, which is equal to 10,000 mg kg−1, dry-weight basis) in sediment cores collected from the South (blue lines and squares, N = 20) and North (red lines and circles, N = 25) Red Sea. Smooth lines were constructed using the adjacent-averaging method using OriginPro 2022 software.

Full size image
Table 1 Average element concentrations of the Red Sea sediment cores.
Full size table
Fig. 2: Scatter plots of element accumulation rates in the Red Sea sediment cores.
figure 2

a–o, Total organic carbon (TOC), total nitrogen (TN), Na, Mg, Si, P, S, Ca, V, Cr, Fe, Ni, Cu, Zn, and Cd element accumulation rates (EAR, μg cm-2 decade-1, dry weight) in sediment cores collected from the South (blue lines and squares, N = 19) and North (red lines and circles, N = 24) Red Sea. The EAR data were standardized and presented in different orders of magnitude for comparison, as noted in each legend. The accumulation rates of TOC, Na, Mg, Si, S, Ca, and Fe were multiplied by a factor ranging from 1 to 250, whereas those of TN, P, V, Cr, Ni, Cu, Zn, and Cd were divided by a factor ranging from 1 to 1/2500. Values below or above the y-axis scale are labeled and highlighted with a black symbol. Smooth lines were constructed using the adjacent-averaging method using OriginPro 2022 software.

Full size image

Although Cd is not commonly known as a required nutrient for organisms, it is a bioactive trace element24,25 that can effectively stimulate the growth of marine phytoplankton in seawater with depleted Zn26. This study found that Cd concentrations in South Red Sea sediment cores were positively correlated with nutrient element concentrations, such as TOC, TN, and P (R values > 0.58, p < 0.01; Fig. 3a), and were higher than in the North Red Sea (Table 1). This finding is consistent with a previous study on the Red Sea zooplankton15, which reported higher nutrient element concentrations (e.g., TOC, TN, and P) in zooplankton samples collected in the South Red Sea compared to the central and northern Red Sea and positive correlations with Cd.

Fig. 3: Heatmaps of Pearson correlations among age, elemental concentrations, and accumulation rates; scatter plot of Mg/Ca ratios vs. age in Red Sea sediment cores.
figure 3

a–c Heatmaps of the Pearson correlation coefficients between chronological age (years, marked as 1870- and 1870+ for before and after the 1870s), 15 elemental concentrations, and d–f 15 element accumulation rates of the sediment cores collected from South, North, and both regions of the Red Sea (RS). *p < 0.05, **p < 0.01. Scatter plot of the linear regression between Mg/Ca concentration ratios and chronological age (years, abbreviated Y in the legend) of the g, North (red lines and circles, N = 25), and h, South (blue lines and squares, N = 20) Red Sea sediment core.

Full size image

The analysis suggests that nutrients in the Red Sea originate from the Indian Ocean, with their concentrations decreasing at a gradient toward the North due to the semi-enclosed structure of the Red Sea, the absence of riverine inputs, and the negligible precipitation/runoff. This finding is supported by previous studies5,12,13 that the nutrient sources of the oligotrophic Red Sea rely primarily on the horizontal intrusion (wind-induced) of nutrient-rich waters from the Indian Ocean through the Gulf of Aden. Furthermore, the data reveal significantly higher trace metal concentrations, such as V, Cr, Fe, Ni, Cu, Zn, and Cd, in the South Red Sea sediment cores (p < 0.05, Table 1). These results suggest that the Indian Ocean is a nutrient supplier and an important source of trace metals to the Red Sea.

Since the onset of the Second Industrial Revolution, the mean accumulation rates of various nutrient elements, including TOC (-14.9%), TN (-14.0%), P (-16.4%), S (-17.2%), and Cd (-23.0%), have notably decreased in the South Red Sea sediment core. Moreover, the decline rate increased significantly after the 1970s, as depicted in Figs. 2 and 3e and Table 2. The linear regression slopes between the TOC, TN, and Cd element accumulation rates and the years before and after 1870 have decreased remarkably by 2.79, 2.93, and 5.54 times, respectively (Table 3). These findings indicate that the Red Sea received fewer nutrients from the Indian Ocean at an accelerating pace after the 1870s and 1970s. The reduction in nutrient supplies may potentially be attributed to the rapid warming and enhanced ocean stratification over the Indian Ocean27,28 and the Red Sea4,5 over the past 150 years, as a result of increasing greenhouse gas emissions worldwide29. Notably, organisms recycle both Cd and P, and the ratio of dissolved Cd/P concentrations in the Indian Ocean was observed to increase with depth30. Our findings indicated a notable reduction in Cd/P concentration ratios in sediments from the South Red Sea post-1800s (Supplementary Fig. 1, Supplementary Data 2). This reduction could imply a more frequent utilization of surface waters by organisms or a diminished upwelling of water from the Indian Ocean into the South Red Sea. Intriguingly, this trend seems to have manifested possibly before the onset of the Second Industrial Revolution (~1870s).

Table 2 Average element accumulation rates (EARs) of the Red Sea sediment cores.
Full size table
Table 3 Linear regression slopes between the element accumulation rates (EAR, μg cm−2 decade−1) and years, and the rate of change in the slope before and after the 1870s for the Red Sea sediment cores.
Full size table

Sediment Mg/Ca concentration ratios have been widely employed to reconstruct past sea surface temperatures (SSTs) in tropical oceans where instrumental data are limited31. As per thermodynamic laws, higher seawater temperatures favor the substitution of Mg for Ca in carbonate sediments32. The current study found a significant change in the Mg/Ca ratio in the South Red Sea during the end of the Little Ice Age, followed by a less steep trend that has continued since the industrial revolution (Fig. 3h). In the North Red Sea, a drastic change occurred in the last 70 years, coinciding with the increase in oil exploitation (local effects) and acceleration of global change effects (Fig. 3g; Supplementary Data 3).

The Mg/Ca concentration ratios in the South Red Sea have a positive correlation with time (R-value = 0.75, p < 0.01, Fig. 3h), indicating a considerable increase in SSTs in the South Red Sea in the past few centuries. This hypothesis align with a previous study33 that global temperatures have steadily risen since the end of the Little Ice Age for 200 years. Furthermore, it is noteworthy that the warm pool areas in the tropical Indian Ocean, characterized by SSTs exceeding 28 °C and holding a direct connection to the South Red Sea through the Gulf of Aden, have undergone a rapid expansion over the last 150 years28.

As global warming continues and sea levels rise, the Indian Ocean may expand, leading to an increased flow of water into the Red Sea. Additionally, the Red Sea’s elevated temperatures, compared to the Indian Ocean, could result in enhanced evaporation rates. This, in turn, might augment the transfer of water from the Indian Ocean to the Red Sea. However, as the primary supplier of nutrients and trace elements to the Red Sea, the Indian Ocean’s rapid warming (and that of the Red Sea) is likely to intensify ocean stratification. This amplified stratification could disrupt the vertical exchanges between the ocean’s surface and its deeper, nutrient-rich layers9,10,27. As a result, the nutrient concentration of the water entering the Red Sea might decrease, leading to a potential decline in nutrient inputs. On the other hand, warming is expected to weaken the monsoon winds over the Red Sea and Indian Ocean34,35, potentially impairing the driving force of horizontal exchange of nutrient-rich intermediate water intrusion from the Indian Ocean into the Red Sea that occurs during the summer southwest monsoon.

Anthropogenic activity-driven pollutant accumulation in the Red Sea

Sediment core analysis has provided valuable insight into the temporal increase in element concentrations and their correlation with human activities in Saudi Arabia. The North Red Sea sediment core displayed a significant rise in most measured element concentrations, except for TN and Ca, from the 1930s onwards, as depicted in Fig. 1. The analysis suggested that this increase coincided with the discovery of Saudi Arabia’s first oil field, the Dammam oilfield, in 1938.

Furthermore, we observed a noteworthy increase in the average element accumulation rates of the North Red Sea sediment cores after the 1870s, with a minimum growth rate of 0.85% for TN and a maximum growth rate of 30.1% for TOC (Figs. 2 and 3f and Table 2). The linear regression analysis revealed that the slopes of certain trace metal accumulation rates, including V, Cr, Fe, Cu, Zn, and Cd, experienced a significant surge after the 1870s. The slopes of these element accumulation rates multiplied by 3.94, 8.00, 3.32, 10.03, 9.05, and 4.10 times, respectively (Table 3). Moreover, this study indicated that all measured element accumulation rates experienced a higher rate of increase after the 1970s (Fig. 2), coinciding with Saudi Arabia’s industrialization along the Red Sea’s eastern coast. This period includes the establishment and operation of the Yanbu Industrial City Zone (1975), the Saudi Aramco Lubricating Oil Refining Company (1976), the Yanbu Cement Company (1977), the Yanbu Commercial Port (expansion in 1979), the Yanbu Industrial Port (1980), the Yanbu Power Plant (1980), the East–West Crude Oil Pipeline (1981), and the Saudi Arabian Oil Company and Mobil Yanbu Refining Company (1981). This finding aligns with the reports by Al-Rousan et al. 14, who noted a substantial increase (ranging from 17% to 300%) in most trace metal concentrations, including Cd, Cu, Mn, Pb, and Zn, in hard corals collected from the North Red Sea since the 1960s.

These data indicate an alarming increase in the accumulation of various elements, including trace metals, in the North Red Sea. This phenomenon is primarily attributed to the rapid expansion of anthropogenic activity in the region over the past 150 years. The shipping activity surge following the opening of the Suez Canal in 186917 and the rapid development of oil-related sectors in Saudi Arabia and Egypt, including oil refineries and petrochemical plants18, notably contributed to this trend. Several other anthropogenic activities have also played a notable role in the accumulation of these elements in the North Red Sea, such as dredging/filling, operation of crude oil terminals/ports, accidental oil spills, deliberate attacks on oil infrastructures36, and the construction and running of high-pollution industries, such as cement factories15 and power plants that burn crude oil as fuel. The rapid urbanization and massive population boom in coastal cities, such as Jeddah, Rabigh, and Yanbu, exacerbated this situation further. The semi-enclosed structure of the North Red Sea basin, absence of river inputs, rare precipitation, and high seawater evaporation rates37,38 limit water exchange in the region, making discharged pollutants more persistent in the aquatic environment. Thus, the biota of the Red Sea is more vulnerable to trace metal pollution.

Red Sea ecosystem prediction

The Red Sea is undergoing rapid warming and facing reduced nutrient supplies from the Indian Ocean in the South. Simultaneously, it is experiencing heightened pollutant levels, including trace metals, from human activities in the North. We propose the term “Cai-Agusti Marine Crisis Conflux (CAMCC)” to describe the confluence of these three stressors. Within this context, it’s evident that these intertwined factors have played a crucial role in sculpting the marine ecosystems of the Red Sea. For instance, ocean warming has caused massive coral bleaching events in the Red Sea6,7 due to thermal stress and accumulation of reactive oxygen species39. The increasing ocean stratification caused by ocean warming could also reduce the supply of limited nutrients to the Red Sea8,10 and further challenge the coral’s capacity to produce sufficient antioxidant enzymes to resist thermal stress40. Moreover, excessively high concentrations of trace metals, such as Cu, can restrain key biological processes in coral and increase energy costs for cellular detoxification and repair41, contributing to reduced resistance of coral to thermal stress under warming scenarios42.

The Cai-Agusti Marine Crisis Conflux (CAMCC) delineates a scenario with intensifying thermal stress, declining nutrient levels, and escalating elemental pollution, especially trace metals. This conflux likely synergizes with other prevailing stressors, such as high ultraviolet B radiation, and newer threats like microplastics. Collectively, these pressures can gravely endanger Red Sea marine life. Potential outcomes include reduced food availability8,10, deterioration of breeding and feeding habitats14,16, amplified species competition, and alterations in marine species abundance, diversity, and distribution43. Ultimately, these combined factors might result in large-scale mortality events44, hasten the extinction of specific species45, and degrade the overall marine ecosystems of the Red Sea. In support of this, the Red Sea corals, in regions where seawater temperatures have risen by about 0.4 to 1 °C since the mid-1970s, exhibited a 30% decrease in skeletal extension and an 18% reduction in calcification rates between 1998 and 201046. Simultaneously, corroborating this trend, there has been a documented shift towards community homogenization, accompanied by a decrease in the average size of coral colonies in the Red Sea, spanning from 1997 to 200947.

In conclusion, the unique environment of the Red Sea presents a rare chance to assess and foresee the implications of global warming and anthropogenic activities on marine ecosystems. The ramifications of these stressors, highlighted by the Cai-Agusti Marine Crisis Conflux (CAMCC) we’ve proposed in this research, could echo well beyond the confines of the Red Sea, influencing marine ecosystems worldwide. Thus, we call for urgent actions to reduce global greenhouse gas emissions, aiming to counteract ocean warming. In addition, we stress the importance of embracing cleaner technologies to minimize pollutants from local anthropogenic activities. Undertaking these proactive initiatives is pivotal to both preserving the Red Sea’s rich biodiversity and preempting analogous ecological challenges in marine environments across the globe.

Methods

Study area and sampling methods

This study collected two sediment cores from the South and North Red Sea. The first core, located at (17.33° N, 41.39° E) and retrieved from a water column depth of 390 m, measures 20 cm in length with an inner diameter of 8.6 cm. It was obtained using a gravity corer from the South Red Sea in November 2016. The second core, situated at (27.56° N, 35.28° E) and derived from a water column depth of 1050 m, has a length of 18 cm and an inner diameter of 8.6 cm. This core was sampled using a box corer from the North Red Sea in October 2016 (Supplementary Fig. 2). Different corers were used for sampling primarily because these cores were retrieved during two separate cruises with close dates. Managing to have the same coring instrument for both expeditions posed logistical challenges. It’s worth mentioning that our sampling stations, positioned in the open Red Sea, lie more than 23 km away from the nearest coastline, ensuring they remain unaffected by direct human activities.

After collection, the sediment cores were sliced into segments 0.5 cm thick using a plastic knife on board the research vessel and stored in plastic bags at -20 °C for preservation. Once in the laboratory, the samples were transferred into 50 mL centrifuge tubes and freeze-dried for at least 24 h to a constant weight. The samples were homogeneously vortexed for laboratory analysis. Clean techniques were strictly followed throughout the sampling and laboratory work to avoid possible metal contamination, including wearing disposable nitrile gloves, performing a two-person sample collection procedure (clean/dirty hands standard operating procedure), and using ultra-clean (acid-washed) sample collection/digestion containers.

Sediment chronology analysis with 210Pb and 14C

Alpha spectrometry was applied to measure the activity concentrations of 210Pb in the upper 0 to 10 cm of the sediment cores. An acid attack using a microwave was performed on the sediment samples48. Subsequently, 210Po was measured, assuming radioactive equilibrium between both radionuclides. The chronological age of each sediment core segment was determined using the constant flux: constant sedimentation model49 and the constant rate of supply model50. We utilized the 210Pb dating technique for surficial sediment cores primarily due to its stability against human-induced alterations. Notably, atomic bomb activities released surplus 14C into the environment, which can distort 14C analysis results for surficial sediments. However, the presence of 210Pb typically wanes around the 10 cm depth in sediments, constraining its use for dating deeper sedimentary sections.

To address this limitation, the nonsurface Red Sea sediment core segments (depth >10 cm) were analyzed using the 14C dating technique. This technique employed accelerator mass spectrometry in the Mini radioCarbon Dating System51. The system, powered by a 200 kV tandem, is commercialized by the Swiss company Ionplus and adheres to the ISO 17025 and 9001 standard procedures52. The raw radiocarbon dates underwent calibration using the R software package “Bacon” for Bayesian chronology building53.

During the realization of this study, analyses of reagent blanks, replicates, and the standard reference material (e.g., International Atomic Energy Agency marine soil samples - IAEA-315 for 210Pb analysis) were conducted in each batch of samples to assess for any contamination and ensure reproducibility of the results.

Total organic carbon and total nitrogen analysis

In this study, the CHNS organic element analyzer (Flash 2000, Thermo Fisher Scientific, USA) was used to analyze the TOC and TN content of the Red Sea sediment samples15. Specifically, 10 mg of freeze-dried samples were placed in silver capsules (Organic Elemental Analysis Labs Company, UK). These samples were then subjected to acid digestion using increasing volumes (10, 20, 30, 40, and 50 μL) of 3 mol L-1 HCl (Fisher Chemical, USA). Each digestion step was conducted at 60 °C for 10 minutes. Prior to analysis, standard curves for TOC and TN concentrations were constructed using 2,5-(Bis(5-tert-butyl-2-benzo-oxazol-2-yl) thiophene (C26H26N2O2S, Thermo Fisher Scientific, USA) at concentrations of 2, 4, 6, 8, and 10 mg, demonstrating a high degree of correlation (R-values > 0.997). Following digestion, the silver capsules for each sample were carefully wrapped into balls and placed in tin capsules (Organic Elemental Analysis Labs Company, UK) for analysis. silver and tin capsule without a sample was wrapped into a ball and analyzed using the instrument as the sample blank.

Element analysis and element accumulation rate calculation

50-mg freeze-dried sediment samples were digested using an acid mixture (6 mL of 69% HNO3 and 2 mL of 30% H2O2, Fisher Chemical, USA) via a microwave digestion system (single reaction chamber microwave digestion system, Milestone Technologies, USA). The samples were first digested at 130 °C and 10 MPa (100 bar, 98.69 standard atmospheric pressure) for 10 min and then redigested at 240 °C and 10 MPa for another 25 min. After cooling, the sample solutions were diluted to 25 mL using Milli-Q distilled water and analyzed using Inductively coupled plasma-optical emission spectrometry (ICP-OES, Model: 5110, Agilent Technologies, USA)54.

The Na, Mg, P, S, and Ca concentrations were measured separately against V, Cr, Fe, Ni, Cu, Zn, and Cd concentrations to mitigate potential systematic error or interference effects. The calibration curves of the measured element standard solutions (0.1, 1, 10, and 100 µg L-1) displayed correlation coefficients (R values) greater than 0.999. The limit of detection, quantitation, and practical quantitation limits for the mentioned elements were less than 0.019, 0.063, and 0.16 mg kg-1, respectively. Three replicates of 50 mg standard reference material (SRM) of inorganics in marine sediment (SRM 2702, National Institute of Standards and Technology, USA) and sediment for solid sampling (SRM 2703) were also analyzed alongside the sediment samples to verify data validity and the precision/accuracy of the analytical methods. A series of quality control procedures were strictly followed: running continuous calibration verification standard solutions, laboratory fortified blanks (same as acid or sample blanks), laboratory reagent blanks (acid blanks spiked with standard element solutions), quality control standards, and duplicated samples spiked with quality control standard solutions to ensure the reliability of the ICP-OES instrument. The recovery rates for all these quality control procedures ranged from 90% to 110%, whereas the measured element concentrations of the standard reference materials were consistent with the reference concentrations (78.91–119.57%; Supplementary Tables 2–3).

The EARs (μg cm-2 decade-1, dry-weight basis) in sediments were used to study environmental changes, reconstruct past conditions, and identify pollution sources55. The EARs at different vertical depths (or different chronological ages) of the Red Sea sediment cores were calculated using the following modified equations56:

$${{{{{\rm{EAR}}}}}}=\, {{{{{\rm{EC}}}}}}\left({{{{{{\rm{\mu }}}}}}{{{{{\rm{g}}}}}}\; {{{{{\rm{g}}}}}}}^{-1}\right)\times {{{{{\rm{Dry}}}}}}\; {{{{{\rm{Bulk}}}}}}\; {{{{{\rm{Density}}}}}}\left({{{{{{\rm{g}}}}}}\; {{{{{\rm{cm}}}}}}}^{-3}\right)\\ \times {{{{{\rm{SR}}}}}}\left({{{{{{\rm{cm}}}}}}\; {{{{{\rm{decade}}}}}}}^{-1}\right)$$
(1)
$${{{{{\rm{SR}}}}}}=\frac{{{{{{\rm{Depth}}}}}}\; {{{{{\rm{at}}}}}}\; {{{{{\rm{year}}}}}}\; {{{{{\rm{Y}}}}}}\left({{{{{\rm{cm}}}}}}\right)-{{{{{\rm{Depth}}}}}}\; {{{{{\rm{at}}}}}}\; {{{{{\rm{year}}}}}}\; {{{{{\rm{X}}}}}}\left({{{{{\rm{cm}}}}}}\right)}{\left({{{{{\rm{Year}}}}}}\; {{{{{\rm{X}}}}}}-{{{{{\rm{Y}}}}}}\right)\times 10}$$
(2)

where EC and SR are the element concentration and sedimentation rate of the Red Sea sediment cores, respectively. The sediment dry bulk density was calculated as the mass of dry solids in a given volume of sediment (Supplementary Data 1). The relative changes of the sediment EAR before and after the 1870s were calculated using the following equation:

$${{{{{\rm{Relative}}}}}}\,{{{{{\rm{changes}}}}}}\,{{{{{\rm{of}}}}}}\,{{{{{\rm{EAR}}}}}}=\frac{[{{{{{{\rm{EAR}}}}}}}_{({{{{{\rm{years}}}}}} > 1870)}-{{{{{{\rm{EAR}}}}}}}_{({{{{{\rm{years}}}}}} < 1870)}]\ast 100 \% }{{{{{{{\rm{EAR}}}}}}}_{({{{{{\rm{years}}}}}} < 1870)}}$$
(3)

To analyze the relationship between sediment EAR and chronological age, a linear regression analysis was employed to determine the slopes of the regression lines. The calculation of the EAR slope change rates before and after the 1870s was conducted using the following equation:

$${{{{{\rm{EAR}}}}}}\,{{{{{\rm{slope}}}}}}\,{{{{{\rm{change}}}}}}\,{{{{{\rm{rate}}}}}}=\frac{[{{{{{\rm{Slope}}}}}}_{({{{{{\rm{years}}}}}} > 1870)}-{{{{{\rm{Slope}}}}}}_{({{{{{\rm{years}}}}}} < 1870)}]\ast 100 \% }{|{{{{{\rm{Slope}}}}}}_{({{{{{\rm{years}}}}}} < {{{{{\rm{1870}}}}}})}|}$$
(4)

where |Slope(years <1870)| is the absolute value of the slope(years <1870).

Silica analysis

The Si sediment concentrations were not measured using ICP-OES due to the inability of nitric acids to dissolve the silicate compounds in the sediment samples. Although hydrofluoric acid can dissolve the silicate compounds of the sediment samples, its use poses considerable safety risks and can cause damage to the glassy components of the ICP-OES instrument (such as the pneumatic nebulizer). Instead, the Si concentrations in the sediment samples were assessed using an X-ray fluorescence spectrometer (Model: XGT-7000, HORIBA Scientific, Canada). 10 mg of 12 sediment samples were weighed and evenly spread on the instrument’s sample holder. The instrument selected four measuring points for each sample and scanned these points for 300 s. After that, the integrated software calculated the peak area and corresponding Si concentration for each measuring point57.

Statistical analysis

The differences in element concentrations between the South and North Red Sea sediment core were analyzed using one-way analysis of variance or two independent sample nonparametric tests in the SPSS statistical analysis software (Version 27.0, IBM Corporation, USA), based on the homogeneity of variance. The linear regression analysis was employed to calculate the slopes of the regression lines between the chronological age and EARs of the Red Sea sediment cores. Multivariate correlation analysis was conducted using OriginPro software (Version 2022, OriginLab Corporation, USA) to investigate Pearson correlations within or between the years (chronological age), element concentrations, and EARs of the sediment cores. Statistical significance was defined as a p value < 0.05.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.