Historical reconstruction of organic carbon inputs to the East China Sea inner shelf: Implications for anthropogenic activities and regional climate variability

Abstract

A gravity core collected from the East China Sea (ECS) inner shelf was analyzed for elemental and stable isotopic composition, lignin-phenols, and sedimentary pigments to investigate changes of organic carbon (OC) inputs during the past two centuries. In particular, we examined the linkages between terrestrial and marine OC inputs with climate variability and anthropogenic activities. The decrease of terrestrial OC contribution (from 41% to 28%) and increasing diagenetic indices of lignin-phenols (P/(S + V): from 0.12 to 0.22; 3,5-Bd/V: from 0.03 to 0.09) after the 1970s were possibly attributed to intensified deforestation, dam construction, and channel erosion. Lignin content (Λ₈) ranged from 0.35 mg/100 mg OC to 6.92 mg/100 mg OC, with lower values corresponding to the worst flooding events in the Changjiang watershed and weaker East Asian Winter Monsoon (EAWM), while higher Λ₈ was more correlated to the strengthening of EAWM. This indicates that terrestrial inputs to Zhe-Min Coast are different from those in Changjiang Estuary during flooding events and strongly linked with regional climate variability. The total contents of sedimentary chloropigments (i.e. pheophorbide-a, pheophytin-a, pyropheophytin-a, sterol chlorin esters, and carotenol chlorin esters) ranged from 663.4 to 74.9 nmol g⁻¹ OC, and decreased exponentially downwards. Sedimentary chloropigments that were used to document historical change of phytoplankton biomass were decoupled with historical changes of Changjiang riverine nutrient inputs but corresponded well to the fluctuation of regional climate variability. Higher phytoplankton biomasses usually were observed during positive phases of Pacific Decadal Oscillation (PDO) and/or warm El Niño-Southern Oscillation (ENSO) events, and lower algal biomass usually corresponded to the negative phase of PDO and/or cold ENSO events.

Keywords

anthropogenic activity chloropigment East China Sea lignin organic carbon regional climate variability

Introduction

Large-river delta-front estuaries are unique coastal environments receiving high inputs from both terrestrial (e.g. nutrients, sediments, and organic matters) and marine (primary production) sources, making them natural ‘recorders’ of anthropogenic activities and regional climate change (Bianchi and Allison, 2009). In past decades, increasing dam construction in river catchments worldwide has significantly reduced riverine sediment supply to the coast, causing coastal erosion, wetland losses, and changes in the composition and abundance of benthic and pelagic algae in large-river delta-front estuary systems (e.g. Bianchi and Allison, 2009; Duan and Bianchi, 2006; Gong et al., 2006; Xu and Milliman, 2009). In fact, coastal eutrophication has become one of the most important issues in coastal management across the globe, which is largely linked with human activity (e.g. fertilizers, sewage) in watersheds (Conley et al., 2009). For example, a disproportional enrichment of nutrients due to increasing anthropogenic activities in river drainage basins has resulted in serious ecological consequences such as harmful algal blooms (HABs) and increased frequency of hypoxia in large-river delta-front estuaries and adjacent shelves (e.g. Caballero-Alfonso et al., 2015; Carpenter, 2005; Lewitus et al., 2012; Li et al., 2011; Savage et al., 2010). Regional climate variability, such as the Pacific Decadal Oscillation (PDO) and El Niño-Southern Oscillation (ENSO), has also affected the coastal and open ocean primary production via variations in sea water temperature, current circulation, and nutrient supply (Barange and Perry, 2009; Chavez et al., 2010; Gregg et al., 2003). The eco-environmental evolution of large-river delta-front estuary systems is a complex function of all the aforementioned drivers, and distinguishing differences between the role of anthropogenic activities and regional climate variability remains a key challenge.

As the largest continental marginal sea in the West Pacific, the East China Sea (ECS) is a region that has suffered from significant anthropogenic influences (Tian et al., 1993). During recent decades, the Changjiang watershed has experienced rapid urbanization with extensive usage of chemical fertilizers and discharge of domestic waste, which resulted in more frequent HABs and seasonal hypoxia in the Changjiang Estuary and adjacent ECS shelf (Dai et al., 2011; Li et al., 2011; Zhao et al., 2012; Zhou et al., 2008). Moreover, more than 50,000 dams (including dams on farmers’ fields and dams higher than 100 m) have been constructed in the Changjiang catchment since the 1950s, which has led to significant decreases of riverine sediment discharge and changes in the aquatic ecosystems (Yang et al., 2005, 2011b; Zhang et al., 2014). Also, there has recently been an increase in the number of studies on historical reconstruction of eco-environmental evolution of the Changjiang Estuary and adjacent ECS shelf, including the history of eutrophication, phytoplankton community changes, hypoxia, Changjiang flooding and drought events, and the response to monsoons (Hu et al., 2014; Jiang et al., 2014; Li et al., 2011, 2013; Liu et al., 2015; Wang et al., 2014a, 2014b; Zhao et al., 2012). However, little remains known about the relative importance of anthropogenic activities and regional climate variability on eco-environmental evolution in different areas of the Changjiang Estuary and adjacent ECS shelf.

The primary objectives of this study were to (1) reconstruct the historical trends of organic carbon (OC) inputs from both terrestrial and marine sources in the ECS inner shelf during the past two centuries, and (2) distinguish the relative importance of regional climate variability and human activities on the eco-environmental evolution of this region. In general, we measured grain size composition, elemental and stable isotopic composition, lignin-phenols, and sedimentary pigments in a ²¹⁰Pb-dated gravity core collected from the Zhe-Min coastal mud area to characterize the basic properties of sediments and OC. A three end-member mixing model based on Monte-Carlo simulation was employed to estimate the relative contribution of OC from different sources. Finally, we collectively examined the linkages between OC inputs and eco-environmental changes in this region for the first time, such as eutrophication, HABs, and the effects of anthropogenic activities and regional climate variability.

Materials and methods

Study area and sample collection

As the longest river (6300 km) in China, the Changjiang originates from the Qinghai-Tibetan Plateau, stretches across 11 provinces (municipalities), and drains into the ECS (Milliman and Farnsworth, 2011; Yang et al., 2011b). Large amounts of riverine terrestrial matters are transported into estuarine and coastal areas by the Changjiang resulting in the formation of coastal mud areas along the ECS inner shelf (Figure 1; Guo et al., 2002; Liu et al., 2006, 2007). The dispersal of the eutrophic Changjiang Diluted Water (CDW) plume is largely dominated by the physical oceanography of this region, which is governed by the seasonal varied Taiwan Warm Current (TWWC), Yellow Sea Coastal Current (YSCC), Zhe-Min Coastal Current (ZMCC), Changjiang River discharge, and tidal cycle (Figure 1; Rong and Li, 2012; Wang et al., 2007). Although characterized by high water discharge, most of the CDW is restricted at the estuarine area due to the barrier and/or shear effect of strong northward TWWC during flood season (Guo et al., 2002). The Kuroshio Subsurface Water (KSW) and TWWC can intrude into the ECS inner shelf and form upwelling due to wind driven current and orographic uplift (Chen et al., 2004; Shi et al., 2013), which has become an important nutrient source to the coastal region (Chen, 2008; Chen and Sheu, 2006). The intrusion of KSW and TWWC to the ECS inner shelf are highly influenced by the shift of North Equatorial Current bifurcation latitude that is dominated by regional climate variabilities (e.g. PDO and ENSO; Chang and Oey, 2012; Wu, 2013). Generally, the Zhe-Min coastal region has a sedimentary environment that suffers from less physical disturbance compared with the shallower Changjiang estuarine area. The Zhe-Min coastal region is also largely influenced by the CDW, ZMCC, TWWC, and KSW, making it a better place to study the relative influence of regional climate variability and anthropogenic activities on the eco-environmental changes in this region.

Figure 1.

Location of Core D2 in the ECS inner shelf. Arrows indicate the direction of the currents (from Liu et al., 2007). Red dot indicates the sample location.

A sediment core was collected onboard R/V Runjiang 1 (Zhoushan Runhe Co., Ltd, China) using a gravity corer at site D2 (122.17° E, 28.50° N, water depth: 40 m) from the Zhe-Min coastal mud area on 3 June 2011 (Figure 1). This core was split in half vertically, and half core was sliced at 2- to 5-cm intervals throughout the core using stainless-steel cutter. Sub-samples were transferred into pre-combusted aluminum boxes and stored at −20°C freezer. The other half-core was sliced at 0.5- to 50-cm intervals and transferred into clean plastic bags for the determination of the particle reactive radionuclide ²¹⁰Pb. In the lab, the first half-core was freeze-dried and stored at −20°C for further analysis.

Sediment chronology

The ²¹⁰Pb activities in core sediments have been widely used to determine the chronology of coastal sediments in this region (e.g. Li et al., 2013; Liu et al., 2007; Zhu et al., 2014). The detailed analytical method for down-core activity of ²¹⁰Pb is provided in Xu et al. (2014). Briefly, about 50-g freeze-dried and homogenized sub-samples were sealed into a Perspex container, and were determined for the total ²¹⁰Pb content at 46.5-keV gamma emission by germanium gamma spectrometer (Canberra Be3830). After 3–4 weeks’ growth of radon daughter, ²¹⁴Pb at 295- and 351-keV and ²¹⁴Bi at 609-keV gamma emissions were counted in order to determine the content of ²²⁶Ra. All samples were counted for at least 24 h in order to assure the counting efficiency and to reduce standard errors below 15%. The excess ²¹⁰Pb activity was calculated by subtracting the activity of ²²⁶Ra from the total ²¹⁰Pb activity. Errors were calculated taking account of counting statistics and propagation with a confidence interval of 1σ. Detector efficiencies for this geometry were calculated using a natural sediment standard (GBW04304) and detector backgrounds were determined using vessel blanks at the energies of interest.

Elemental and stable carbon isotope

The detailed analytical methods for elemental and stable carbon isotope composition are described in Yao et al. (2014). Briefly, approximately 50 mg of homogenized dry sediments were fumigated with hydrochloric acid vapor to remove the carbonate fraction prior to instrumental analysis. This fumigation method has been shown to be efficient in decarbonation and suitable for determining the OC content and stable carbon isotope ratio (δ¹³C) in coastal sediments with low carbonate (Komada et al., 2008; Wang et al., 2015), and has been widely used in previous studies of ECS inner shelf (Li et al., 2011, 2012, 2013, 2014a; Wang et al., 2015). Total organic carbon (TOC) and stable carbon isotope ratios were then analyzed using a Carlo Erba NA 1500 Series 2 nitrogen/carbon analyzer (Fisons Instruments, USA) interfaced to a DELTA PLUS^XP continuous flow isotope ratio mass spectrometer (Thermo Finnigan Instruments, USA). Analytical precision of TOC was ±0.02 wt% (n = 6). δ¹³C is expressed in δ notation in per mil relative to the international standards Vienna Pee Dee Belemnite. The precision based on replicatemeasurements for δ¹³C was better than ±0.1‰ (n = 6).

Grain size composition and specific surface area

Grain size analysis was conducted using a laser particle size analyzer (Mastersizer 2000; Malven Instruments Ltd, UK) after the method of Hu et al. (2009b). The relative standard deviation of replicate samples was less than 3% (n = 6). Particles less than 4 µm in size were classified as clay, 4–63 µm as silt, and those greater than 63 µm as sand. The sediment specific surface area (SSA) was measured based on the method reported by Waterson and Canuel (2008), and the detailed analytical steps are provided in Yao et al. (2014). Briefly, about 1 g of freeze-dried sediment was heated at 350°C for 12 h in order to remove organic matter using muffle furnace. Then, the SSA of organic matter–free sediment was determined by nitrogen adsorption according to 5 point BET method using automatic surface area analyzer (3H-2000BET-A; Beishide Instrument-ST Co., Ltd, China).

Sedimentary pigments

Sedimentary pigments were extracted according to the method of Zhao et al. (2012). Briefly, approximately 2-g homogenized freeze-dried sediments were weighed into a centrifuge tube containing 6 mL acetone. The tube was vibrated by vortex stirrer, ultrasonicated in an ice bath for 1 min, and then stored at −20°C for 12 h. The extract was separated from sediment by centrifugation (5 min, 5000 r min⁻¹, 4°C), and the supernatant was transferred to a large glass tube. After that, the extraction procedure was repeated several times, without storage, until the supernatant was colorless. The extracts were combined together, dried under gentle N₂ stream, re-dissolved with 100 µL acetone, and filtered with a syringe filter (0.45 µm PTFE; Whatman, UK) before injection. The whole extraction processes were performed under low light conditions.

Pigments were analyzed using high performance liquid chromatography based on the method of Wright et al. (1991), as modified by Chen et al. (2001). Briefly, 50 µL of the extract was injected into a Waters 610 solvent delivery system coupled with a Waters 996 photodiode array detector and a Shimadzu RF 535 fluorescence detector (excitation set at 440 nm and emission set at 660 nm, for identification purposes only). The separations were performed using a reverse phase C₁₈ column (5 µm, 250 mm × 4.6 mm i.d.; Alltech Adsorbsphere) and the gradient program (1 mL min⁻¹) described by Chen et al. (2001).

The qualitative and quantitative examinations of pigments were based on the retention time, absorption spectra, and area of the peaks in each sample chromatogram (at 666 nm) comparing with those of authentic standards (chlorophyll-a, chlorophyllide-a, divinyl chlorophyll-a, pheophytin-a, and pheophorbide-a, purchased from DHI Water & Environment, Denmark) and literature values (Chen et al., 2005; Zhao et al., 2012). The degradation products of chlorophyll-a without commercial standards (e.g. pyropheophytin-a, sterol chlorin esters, and carotenol chlorin esters) were quantitatively examined based on the response factor of pheophytin-a (Zhao et al., 2012). The precision (coefficient of variation) for measured pigments of interest was better than 3% (Chen et al., 2001). The detection limit of this high performance liquid chromatography instrument was about 1 nM g⁻¹ OC if assuming an OC content of about 1% sediment dry weight (dw; Li et al., 2011).

Degradation model

The early diagenetic degradation processes of organic matter (e.g. sedimentary pigments) are a function of time (Chen et al., 2005; Sun et al., 1991; Zimmerman and Canuel, 2000), and the first-order G-model has been widely used to model the early dia-genesis of sedimentary organic matter in the environments with steady-state mixing and sedimentation rates (Chen et al., 2005; Fujii et al., 2002; Zhu et al., 2014). It has been shown that a fraction of this organic matter is closely associated with mineral matrices, which protect it from being degraded (Stephens et al., 1997; Sun et al., 1991). Since the use of a single degradation rate constant is clearly an oversimplification of decay dynamics in this large-river delta-front estuary, we used a two component equation that resulted in an improvement in fitting the organic matter profiles in most cases (Stephens et al., 1997)

Multi-G model : C_{t} = C_{\infty} + C_{0} * (1 - a) * e^{(- k_{1} * t)} + C_{0} * a * e^{(- k_{2} * t)}

(1)

where C₀ (nmol g⁻¹ OC) is the content of organic matter in surface sediment, C_∞ (nmol g⁻¹ OC) is the asymptotic organic matter content at the infinite depth, and C_t (nmol g⁻¹ OC) is the organic matter content at certain time t (years), k₁ (yr⁻¹) and k₂ (yr⁻¹) are the degradation rate of different parts of organic matter, a (0 < a < 1) is the percentage of refractory portion in organic matter, d (cm) is the sediment layer depth, s (cm yr⁻¹) represents the sedimentation rate, and t can be expressed as the quotient of d divided by s.

The multi-G diagenetic model of decomposition is usually established within the steady-state profile of sedimentary OC; however, an ideal steady-state distribution of sedimentary OC is hard to find. Therefore, when the time-scales of steady-state profiles are much longer than these of OC inputs’ changing period and/or degradation half-life of sedimentary OC, the OC profile can be seen as steady-state distribution (Stephens et al., 1997). The remineralization of metabolized organic matter is assumed to be at a constant k at any time, and the organic matter was regarded as two portions with different reactivities (Stephens et al., 1997; Zimmerman and Canuel, 2000). As one part of sedimentary organic matter pool, the preserved chloropigments usually follow this equation and exhibited exponential degradation trend in the sediment profile (Chen et al., 2005; Stephens et al., 1997; Zimmerman and Canuel, 2000). In this study, we defined the term ‘residua’ as the difference between measured contents and modeled values of sedimentary chloropigments at certain depth. The sum of residua of all sedimentary chloropigments, that is, the net change of chloropigments is assumed to reflect the historical change of phytoplankton biomass in the water column in this region. Based on the work by Zhao et al. (2012), where certain stable chloropigments were integrated over time, this assumption seems reasonable for an estimate of a ‘rough’ index of past phytoplankton inputs. Negative residua can occur if the phytoplankton biomass was lower than that of modern inputs (or surface interval), and positive residua indicates higher phytoplankton biomass.

Lignin-phenols

Lignin-phenol analyses were conducted based on the alkaline CuO method of Hedges and Ertel (1982), as modified by Bianchi et al. (2002), and the detailed analytical steps are provided in Li et al. (2014a). Briefly, appropriately 1 g of homogeneous freeze-dried sediments containing about 4 mg OC was placed into stainless-steel reaction vessel with 330 ± 4 mg CuO and about 4 mL 2 mol L⁻¹ NaOH under nitrogen atmosphere. Then the vessels were heated at 150°C for 3 h in an oven. After that, 50 µL ethyl vanillin was added to each reaction vessel as internal standard before the aqueous reaction products were separated from sediments. Oxidation products were derivatized with bis-(trimethylsilyl)-trifluoroacetamide, and then the analysis of derivatives was conducted on an Agilent 7890A GC with an Equity-5 capillary column (30 m, 0.25 mm i.d., 0.25-µm film thickness; SUPELCO). The quantification of lignin-phenols was based on the recovery of internal standard ethyl vanillin and external calibration curve of lignin-phenol standards containing known amounts of compounds of interest.

A total of 12 compounds were quantified and used as indicators for tracing the sources and diagenetic state of terrestrial vascular plant tissue: V (vanillin, acetovanillone, and vanillic acid), S (syringaldehyde, acetosyringone, and syringic acid), C (p-coumaric acid and ferulic acid), P (p-hydroxybenzaldehyde, p-hydroxyacetophenol, p-hydroxybenzoic acid), and 3,5-dihydroxybenzoic acid (3,5-Bd; Goñi and Hedges, 1995; Hedges and Mann, 1979). The relative standard deviations for the individual compound and the sum of lignin-phenols were less than 8% based on replicate analyses (Li et al., 2014a). The OC normalizedlignin-phenols (Λ₈) are defined as the sum of C, S, and V phenols normalized to 100 mg OC. Degradation/humification state of lignin tissues can be characterized by the usage of different diagenetic indexes: the ratios of P phenol to the sum of S and V phenols (P/(S + V)) and the ratio of 3,5-Bd to V (3,5-Bd/V; Houel et al., 2006; Louchouarn et al., 1999).

Mixing model and Monte-Carlo simulation

Various end-member mixing models have been successfully used to study the variation of different sources of OC inputs and paleoenvironmental changes at the centenary even millennium timescale in aquatic systems around the world (e.g. Carreira et al., 2002; Hu et al., 2009a, 2014; Yang et al., 2011a). Here, we employed a three end-member mixing model using δ¹³C and Λ₈ as source markers and based on Monte-Carlo simulation strategy to distinguish the relative contribution of three different sources of OC in the ECS inner shelf sediments: marine (OC_mar), soil-derived (OC_soil), and vascular plant–derived (OC_VP)

f_{mar} \times δ^{13} C_{mar} + f_{soil} \times δ^{13} C_{soil} + f_{vp} \times δ^{13} C_{vp} = δ^{13} C_{sample}

(2)

f_{mar} \times Λ_{8, mar} + f_{soil} \times Λ_{8, soil} + f_{vp} \times Λ_{8, vp} = Λ_{sample}

(3)

f_{mar} + f_{soil} + f_{vp} = 1

(4)

where f_mar, f_soil, and f_VP are the fractions of OC_mar, OC_soil, and OC_VP in sedimentary OC, respectively. Another important assumption here is that end-member values have not changed significantly over the period of time examined in this study. Although the Changjiang watershed has suffered from certain degree of deforestation, agriculture, and erosion during recent decades, the old age of the riverine particular OC (around 800–1060 years old) and sedimentary OC in the ECS inner shelf (around 2000–5000 years old; Li et al., 2012; Wang et al., 2012) probably indicated that the influence of recent changes on isotopic composition of terrestrial OC would be very limited. Additionally, although the selection or changes of end-member values would result in some uncertainty in thecalculated contributions of different components, the variation trends of relative contributions of different sources of OC remained to be convincing to interpret the environmental changes (Hu et al., 2014).

A detailed description of the mixing model and the Monte-Carlo simulation strategy is provided by Li et al. (2014a). The end-member values of δ¹³C for OC_mar, OC_soil, and OC_VP are −20.0‰ ± 1.0‰, –26.3‰ ± 2.96‰, and −28.1‰ ± 1.68‰, and Λ₈ for OC_mar, OC_soil, and OC_VP are 0 mg/100 mg OC, 1.64 ± 0.89 mg/100 mg OC, and 6.00 ± 5.22 mg/100 mg OC, respectively (Li et al., 2014a). The Monte-Carlo simulation was performed in MATLAB (version R2013a; Math Works, USA) based on a modified script by August Andersson (2011). The mean relative contributions and standard deviations of the three OC sources are calculated for each sample from the Monte-Carlosimulation solutions.

Statistical analysis

Relationships between the measured parameters were determined using SPSS 20 by performing Pearson correlation analysis with a two-tailed test of significance. The program SPSS was also used to run the one-way analysis of variance (ANOVA) to examine the significance of differences in data between two or more groups.

Results

Sediment chronology

A constant initial concentration (CIC) model was used to estimate the sediment accumulation rate in our study. In Core D2, the content of ²¹⁰Pb_excess activities generally showed an exponentially decreasing trend throughout the whole core (R² = 0.84) as the typical ²¹⁰Pb_excess profiles in the CIC model, although there were some ²¹⁰Pb_excess activity fluctuations (Figure 2a; Meng et al., 2015). The sediment accumulation rate indicated by ²¹⁰Pb_excess activities is 1.05 cm yr⁻¹, which is close to the values reported by other studies from nearby regions (e.g. Gao, 2002; Huh and Su, 1999; Liu et al., 2006, 2007; Xia et al., 1999; Yu et al., 2012; Table A.1 and Figure A.1, available online). In fact, an apparently faster decrease in ²¹⁰Pb_excess activities did exist at 40 cm below. If we separately calculated the sediment accumulation rates of the two sections with 40-cm depth as the dividing line, similar sediment accumulation rates were observed with 1.13 cm yr⁻¹ for upper 40 cm and 1.05 cm yr⁻¹ for the whole core. The data of ²¹⁰Pb_excess activities for lower section (below 40 cm) were rather limited, which was insufficient for further age estimation. Here, the CIC model was used for whole Core D2, which provided a 225-year historical reconstruction (from AD 1785 to 2011).

Figure 2.

Vertical profiles of (a) ²¹⁰Pb_excess (black squares), (b) median grain size (red dots), (c) %TOC (blue squares), (d) total organic carbon/specific surface area (TOC/SSA, pink squares), and (e) δ¹³C‰ (green triangles) in Core D2.

Grain size composition and SSA

In general, the fine-grained sediments, silt (from 68.2% to 86.9%, avg. 76.8% ± 3.9%) and clay (from 10.0% to 30.4%, avg. 20.1% ± 4.8%), dominated the sediments in Core D2, with sand (from 0.6% to 9.9%, avg. 3.1% ± 2.0%) contributing the least (Figure A.2, available online; Meng et al., 2015). The median grain size ranged from 7.0 to 21.0 µm with an average of 13.3 ± 3.3 µm (Figure 2b). SSA ranged from 7.5 to 21.7 m² g⁻¹ with an average of 17.3 ± 2.9 m² g⁻¹ (Meng et al., 2015).

TOC, TOC/SSA, and stable carbon isotope

TOC contents ranged from 0.23% to 0.56% (avg. 0.43% ± 0.08%), and showed no clear trends in variation, with some obvious fluctuations around certain years (e.g. 1830, 1930, 1970, 1990, and 2000; Figure 2c; Meng et al., 2015). TOC/SSA ratios ranged from 0.16 to 0.33 mg OC m⁻² with an average of 0.25 ± 0.04 mg OC m⁻², and were pretty stable throughout the core (Figure 2d). The δ¹³C ranged from −23.3‰ to −21.6‰ with a mean of −22.7‰ ± 0.4‰, and generally increased after 1970 with a fluctuation around 2000 (Figure 2e; Meng et al., 2015).

Sedimentary pigments

Chlorophyll-a was not detected in all samples, while the derivatives of chlorophyll-a, such as pheophorbide-a, pheophytin-a, pyropheophytin-a, sterol chlorin esters, and carotenol chlorin esters, were abundant throughout the core (Figure A.3, available online). All chloropigments showed high concentrations in the surface layer, and decreased sharply to a very low content at around 50-cm depth, remained relatively constant with depth (Figure A.3, available online), and all chloropigments showed significant correlations with each other (p < 0.001). Sterol chlorin esters had the highest surface concentrations among all chloropigments and ranged from 175.8 to 19.4 nmol g⁻¹ OC (avg. 49.4 ± 32.4 nmol g⁻¹ OC). Pyropheophytin-a had the highest mean content of 58.9 ± 27.1 nmol g⁻¹ OC (from 166.9 to 29.7 nmol g⁻¹ OC). Pheophytin-a varied from 169.9 nmol g⁻¹ OC in the surface to 10.9 nmol g⁻¹ OC in the bottom of the core with an average of 34.6 ± 30.8 nmol g⁻¹ OC. The contents of pheophorbide-a and carotenol chlorin esters were much lower than pyropheophytin-a, pheophytin-a, and sterol chlorin esters, and varied from 94.6 to 4.2 nmol g⁻¹ OC (avg. 21.0 ± 18.9 nmol g⁻¹ OC) and 63.2 to 1.9 nmol g⁻¹ OC (avg. 8.8 ± 10.6 nmol g⁻¹ OC), respectively.

Based on our degradation model, the half-lives of different chloropigments ranged from several years to hundreds of years, and had a good fit with our model (R² > 0.84; Table A.2, available online). This indicated that the vertical profiles of chloropigments can be viewed as having a steady-state distribution over the past 225 years (Stephens et al., 1997), something that would not be expected closer to the highly dynamic Changjiang LDE. The residua of each chloropigment showed significant correlations with each other (p < 0.05), and showed higher values around the yearsof 1910s, 1920s, and 2000s, and lower value around the years of 1990s and 1890s. No obvious changing trends were observed for the residua of all chloropigments (Figure A.4, available online). As mentioned earlier, we mainly focused on the variation of the sum of residua of all chloropigments that were hypothesized to represent a rough index of total algal biomass in the water column.

Lignin-phenols

The OC normalized lignin-phenols (Λ₈) ranged from 0.35 to 6.92 mg/100 mg OC (avg. 1.49 ± 0.94 mg/100 mg OC), with the highest value detected at a layer corresponding to 1933/1934, and decreasing trends after the 1980s (Figure 3a). The absolute content of lignin-phenols (Σ₈) ranged from 0.01 to 0.33 mg g⁻¹ dw with an average of 0.07 ± 0.05 mg g⁻¹ dw, and showed similar variation trend with that of Λ₈ (Figure 3b). The P/(S + V) ratios ranged from 0.06 to 0.26 (avg. 0.13 ± 0.03) and showed a decreasing trend with increasing depth (Figure 3c); the 3,5-Bd/V showed similar trends with P/(S + V; Figure 3d).

Figure 3.

Variations of (a) Λ₈ (mg/100 mg OC), (b) Σ₈ (mg g⁻¹ dw), (c) P/(S + V), (d) 3,5-Bd/V, (e) peak discharge during flooding events in the Changjiang watershed, and (f) EAWM index in Core D2. The data of peak discharge during flooding events were collected from Shi et al. (2004), Li et al. (2013), and Meng et al. (2015). The gray dotted lines represent the years of the worst flooding events in the Changjiang catchment. The EAWM index data were collected from D’Arrigo et al. (2005).

Three end-member mixing model

Monte-Carlo simulation results of a three end-member model showed that marine OC was the dominant component of sedimentary OC in Core D2, and varied from 41% to 72% with a mean of 58% ± 6% (Figure 4a). The contribution from soil-derived OC ranged from 13% to 33% with an average of 28% ± 4% (Figure 4b). By contrast, the terrestrial vascular plant–derived OC contributed least to the fraction of sedimentary OC (from 7% to 47%, avg. 14% ± 6%; Figure 4c). Generally, the relative contributions from different sources were generally stable before the 1970s, except for higher contributions from vascular plant–derived OC (47%) during years 1933/1934 (Figure 4). After the 1970s, the contribution of terrestrial OC showed a decreasing trend while the relative contributions from marine OC increased (Figure 4b and c). The marine OC content ( ${OC}_{mar}^{'}$ = f_mar × %TOC), soil-derived OC content ( ${OC}_{soil}^{'}$ = f_soil × %TOC), and vascular plant–derived OC content ( ${OC}_{VP}^{'}$ = f_VP × %TOC) generally showed similar varying trend with those of OC fractions (Figure 4d–f).

Figure 4.

Variations of OC fractions from (a) marine (f_mar (%)), (b) soil (f_soil (%)), and (c) vascular plant (f_VP (%)) and different source OC contents from (d) marine ( ${OC}_{mar}^{'}$ ), (e) soil ( ${OC}_{soil}^{'}$ ), and (f) vascular plant ( ${OC}_{VP}^{'}$ ) as determined by Monte-Carlo Simulation in Core D2.

Discussion

Sedimentary environment

The profiles of ²¹⁰Pb, grain size composition, and chloropigments all indicated a relatively uniform sedimentary environment in the region. A significant correlation between ²¹⁰Pb activities and sediment depth (R² = 0.84, p < 0.01), in the absence of surface mixing (Figure 2a), indicated that physical and/or biological disturbances were minor in this core, once again unlike those found closer to the Changjiang Estuary (Duan et al., 2005). The increasing frequency and extent of serious hypoxia events during summer in recent years in the Changjiang Estuary region and adjacent ECS inner shelf (Wang et al., 2012; Zhu et al., 2011) may have inhibited the widespread presence and bioturbation activity of benthic fauna (Schüller et al., 2014); however, this remains purely speculative. Relatively low median grain sizes at this site (avg. 13.3 ± 3.4 µm) and the dominance of fine-grained particles (silt and clay) of sediment compositions also indicated lower energy hydrodynamic conditions (Figure 2b; and Figure A.2, available online; Li et al., 2014a). Additionally, the good model fit strongly indicated that the decay processes of sedimentary chloropigments closely followed first-order decay dynamics; therefore, the supply of phytoplankton chloropigments, and diagenetic processes generally played a more important role than physical mixing in this core (Chen et al., 2005). Exponential losses of sedimentary chloropigments without any obvious change in grain size composition and median grain size throughout this core further implied diagenetic loss of sedimentary chloropigments during progressive burial under steady-state sedimentation (Mayer et al., 2007; Figure 2b; and Figures A.2 and A.3, available online). TOC/SSA ratios have been shown to remain stable for sediments withsimilar sources and have proved to be an effective parameter indicating efficient OC remineralization due to post-depositional disturbances of sediments (Blair and Aller, 2012; Yao et al., 2014). TOC/SSA ratios were very stable throughout Core D2 without obvious fluctuations (Figure 2d), indicating limited post-depositional disturbances.

Although the Changjiang watershed has suffered from severe land-use change and dam construction, its influence on the sedimentation rate in the Zhe-Min coastal mud area, south of the Changjiang Estuary proper, has been limited. This is surprising when considering such dramatic alterations in the Changjiang watershed such as the Three Gorges Dam, which we would have expected to significantly reduce the flux of coarse riverine particles to this region, even when considering the more limited trapping effect on the fine-grained sediments (Xu et al., 2011). The relatively stable and small median grain size (around 13.3 µm) throughout Core D2 further supports any clear effects of the Three Gorges Dam in this region. Second, due to the strong sediment carrying capacity of the Changjiang River, riverbank and delta region downstream of the Three Gorges Dam have been significantly eroded (Yang et al., 2011b). For example, sediments supplied from the river channel to the suspended particle load are around 60–70 Mt yr⁻¹ after the operation of Three Gorges Dam (Luo et al., 2012; Xu and Milliman, 2009). Similarly, the delta front has also devolved from around 125 Mm³ yr⁻¹ of sediment accumulation in the 1960s to around 100 Mm³ yr⁻¹ of erosion in recent years (Yang et al., 2011b). The increasing median grain size of riverbed sediments and decreasing median grain size of suspended particles, after the operation of Three Gorges Dam, are further support for the role of resuspension of sedimentary fine particles (Chu et al., 2007; Xu and Milliman, 2009; Xu et al., 2011). These resuspended fine sediments are probably important supply sources for the riverine sediment discharge and occupy a significant proportion of riverine suspended sediments, particularly after the operation of Three Gorges Dam. Third, the sand–mud boundary in the ECS showed a landward retreating trend (around tens of kilometers) in the Changjiang River mouth region but has expanded seaward in the Zhe-Min coastal area (Luo et al., 2012). This is likely due in part to the increasing proportion of finer grain-sized particles in Changjiang riverine particles. Although the influence of anthropogenic activities on sedimentation rate in this region was probably limited, the terrestrial OC character recorded here indicated remarkable changes during past decades (more details can be found in the next section).

Overall, we found this site to be an excellent location for recording both inputs from the drainage basin and marine primary production, and for testing the relative importance of anthropogenic activities and regional climate variability on eco-environmental evolution in the Changjiang Estuary and adjacent ECS shelf.

Human activity, flooding events, and climate variability effects on terrestrial OC inputs

Human activities (e.g. land-use changes, deforestation, and dam construction) in the Changjiang drainage basin have significantly changed the character of riverine terrestrially derived materials over past decades. As mentioned above, although the damming activity has limited influence on the sedimentation rate in the Zhe-Min coastal area, its influence on the terrestrial OC character remains compelling. The decreasing trend of lignin content and terrestrial OC fractions after the 1970s (Figures 3a, b, 4b, and c) in Core D2 is likely attributed to more river channel and delta erosion accompanied by greater resuspension of fine-grained riverbed sediments (Luo et al., 2012; Milliman and Farnsworth, 2011; Xu et al., 2011; Yang et al., 2011b). This appears to have resulted in the transport of more lignin-poor fine-grained terrestrial OC from deeper soils, due to more intense erosion in the drainage basin, to the Zhe-Min Coast. Higher P/(S + V) and 3,5-Bd/V ratios and lower lignin concentrations after the 1970s further support this notion of sources from deeper soil horizons (Figure 3c and d). For example, Heim and Schmidt (2007) reported that lignin decomposition in soils was very fast and generally showed higher diagenetic state and lower lignin content with increasing depth.

Records of terrestrial OC inputs to coastal sediments, after flooding events in the Changjiang watershed, were remarkably different between the Zhe-Min coastal mud area and the Changjiang estuarine region. For example, lignin parameters in Core D2 varied around flooding years, especially when the worst floods occurred in the Changjiang catchment (e.g. 1954 and 1998; Shi et al., 2004; Figure 3). However, more terrestrially derived materials with very different soil-derived OC character (e.g. lower lignin content and higher degradation state) were delivered to the Zhe-Min Coast during floods (Figure 3), in contrast to more lignin-rich fresher terrestrial-derived OC to the Changjiang Estuary (Li et al., 2013). Hydrodynamic sorting of soil OC and vascular plant debris during transport across the Changjiang Estuary and adjacent ECS inner shelf may explain for these differences (Li et al., 2012, 2014a; Zhu et al., 2013). For example, the Λ₈ value (0.35 mg/100 mg OC) in year 1998 (which corresponded to a worst flood in the entire basin) in Core D2 was relatively lower than that deposited in the Changjiang Estuary (around 0.74–0.95 mg/100 mg OC; Li et al., 2011). Also, the Λ₈ value (0.35 mg/100 mg OC) was much lower than that of surface soils in the Changjiang catchment (around 0.59–3.26 mg/100 mg OC; Yu et al., 2007), which again may be suggestive of a more prominent terrestrial OC contribution from deeper soils. This is consistent with the older radiocarbon age of fluvially derived inputs during flooding events (Li et al., 2012). Such extreme flooding events that occurred in the entire Changjiang watershed would have likely delivered more terrestrially derived materials from upstream with lower Λ₈ values and higher diagenetic state compared with those from midstream and/or downstream (Yu, 2007). Although there is no doubt that more terrestrial OC (including both coarse and fine particles) would be delivered into estuarine areas during flooding events (Li et al., 2013), most of lignin-rich coarse detrital particles would be expected to be deposited in the lower river and delta region, with more soil-derived lignin-poor finer particles being transported further south along the coast (Li et al., 2014a). However, not all flooding events in the Changjiang drainage basin were recorded by organic proxies in Core D2 (Figure 3). Lignin has been shown to be more reactive than previously thought, and diagenetic processes in sediments may obscure the terrestrial OC signals of ancient flooding events (Eadie et al., 1994; Ward et al., 2013). Nevertheless, we did not find significant decrease of TOC/SSA ratios and terrestrial OC contents during ancient flooding events, which probably indicated limited influence of digenetic processes on the terrestrial OC signal of flooding events. The addition of marine OC has been shown to affect the bulk OC signatures in these coastal sediments (Liu et al., 2010), which would probably also mess the difference of terrestrial OC between flooding events and normal period. Moreover, the long distance from Changjiang River mouth may also enhance the temporal decoupling between riverine sediment flux and sedimentary records (Li et al., 2013).

The greater further dispersal and transport of the Changjiang riverine fine sediments in the ECS inner shelf is more dominated by the coastal hydrodynamic conditions (Li et al., 2014a), with the Zhe-Min Coast more highly linked with regional climate variability. Former studies have demonstrated that the deposited fine-grained sediments (<65.5 µm) in the Zhe-Min Coast were mainly derived from the resuspended materials in the estuarine region during winter periods (Guo et al., 2002). Moreover, these fine-grained sediments are more susceptible to the variations of coastal currents affected by the East Asian Winter Monsoon (EAWM; Liu et al., 2010). The lignin content generally showed consistent trends with EAWM (Figure 3a, b, and f). For example, the long period of strong EAWM from the 1920s to 1940s (D’Arrigo et al., 2005) corresponded to the enhanced alongshore transport of Changjiang-derived terrestrial OC and higher Λ₈ value in the Zhe-Min Coast (Figure 3a, b, and f). Conversely, a generally decreasing EAWM since the 1930s may have contributed to the slight decreasing of Λ₈ values (Figure 3a, b, and f). However, if we calculate the correlation between lignin content and strength of EAWM during the same year or adjacent years (±5 year), a fairly weak correlation was observed (R = 0.37, p < 0.05, n = 32).

The most likely reasons for weak statistical linkages between our proxy and regional climate data are the multitude of other anthropogenic changes that have occurred in the Changjiang River watershed. Interestingly, there are many factors which may have contributed to this weak correlation, such as the changes of Changjiang water discharge, dam construction in the Changjiang watershed, sectioning precision, and possible errors of core chronology (Meng et al., 2015). Thus, it is not surprising that there was no significant correlation between lignin-phenol parameters in Core D2 and strength of EAWM. However, we did find that lignin parameters corresponded well to the strength of EAWM (Figure 3). For example, during the strongest EAWM period (around the 1930s), the highest lignin content (Λ₈: 6.92 mg/100 mg OC) and lowest degradation parameter values (P/(S + V): 0.06; 3,5-Bd/V: 0.03) were observed. Conversely, during the weakest EAWM period (around the 1990s), the lowest lignin content (Λ₈: 0.35 mg/100 mg OC) and highest degradation parameter values (P/(S + V): 0.26; 3,5-Bd/V: 0.09) occurred. Moreover, during the periods from the 1780s to 1830s, 1850s to 1910s, 1920s to 1970s, and 1980s to 2000s, the Λ₈ and lignin decay parameters generally showed similar and opposite variation trends with strength of EAWM, respectively. Thus, we conclude that variations of lignin parameters were highly dominated by the strength of EAWM.

Response of phytoplankton biomass to regional climate variability

The intrusions of KSW and TWWC into the ECS inner shelf are key in controlling the historical change of phytoplankton biomass on the Zhe-Min Coast. Although there are no long-term HAB records, the sedimentary records of chloropigment residua corresponded well to the occurrence of HABs in the coastal area of ECS inner shelf after the 1970s (Figure 5). Similarly, the variation of ${OC}_{mar}^{'}$ also exhibited consistent trends with those of residua in most cases (Figure 5), further supporting the possibility of using residua in sediments as a past record of phytoplankton biomass in the water column. Most previous studies have attributed the eutrophication and/or HABs of the Changjiang Estuary and adjacent ECS shelf to the Changjiang inputs (Li et al., 2014b; Zhao et al., 2012; Zhu et al., 2014). While this may be true for the Changjiang Estuary, it may not be the case for the Zhe-Min Coast. As mentioned earlier, most of the CDW was restricted at the estuarine and shelf interface due to the barrier and/or shear effect of the strong northward TWWC (particularly between May and October; Guo et al., 2002); this allowed for enhanced phytoplankton biomass in the Changjiang Estuary during this period (Li et al., 2014b). However, a decoupling between the historical changes of nutrient inputs by the Changjiang (Li et al., 2014b; Jiang et al., 2014) and variations of chloropigment residua in Core D2 (Figure A.5, available online) likely suggests that Changjiang-derived nutrients had only a limited influence on past phytoplankton biomass in the Zhe-Min Coast.

Figure 5.

Variations of sum of residua of all chloropigments (pink bars), ${OC}_{mar}^{'}$ (black solid line), frequency (red circles), and disaster area (blue triangles) of harmful algal blooms (HABs). The data about occurring frequency and disaster area of HABs were collected from China Oceanic Information Network ( http://www.coi.gov.cn/gongbao/zaihai/ ).

Recent work has suggested that the Zhe-Min coastal area has become phosphate-limited for phytoplankton (Shi et al., 2013). It has been shown that most of the phosphates delivered by the Changjiang are quickly consumed by phytoplankton in the CDW plume (varying from >1 µmol L⁻¹ in the river mouth to <0.2 µmol L⁻¹ at the CDW plume front; Chen, 2008), which would obviously limit the effect of CDW on the phytoplankton growth in the Zhe-Min Coast. In contrast, the total annual dissolved phosphate flux from KSW and TWWC to the ECS (around 2.7 × 10¹⁰ moles) is almost two orders of magnitude higher than the total annual riverine input to the ECS (around 3.0 × 10⁸ moles) and ca. 30 times higher than the annual phosphate flux from ECS sediments to the water column (around 7.9 × 10⁸ moles; Zhang et al., 2007). A significant release of potentially bio-available particulate phosphorus from terrestrial particles to the water column with increasing salinity is another possible important source for the dissolved phosphate in the Changjiang Estuary and adjacent ECS inner shelf. However, its maximum annual particulate phosphorous flux (around 6.3 × 10⁷ moles, roughly calculated from He et al. (2009)) is considerably lower than from the KSW and TWWC, and thus not likely as important. Therefore, the intrusions of phosphate-rich KSW and TWWC appear to have dominated the evolution of phytoplankton biomass in the Zhe-Min Coast, in contrast to inputs from the Changjiang and/or local biogeochemical processes.

Given that regional climate variability is closely linked with the KSW and TWWC, we here postulate that they are also significantly linked with phytoplankton biomass on the Zhe-Min Coast. For example, during the positive phase of PDO and warm ENSO events (e.g. 1940 and 2005), the chloropigment residua showed a positive value, indicating a higher phytoplankton biomass compared with normal years (Figure 6). In contrast, during the negative phase of PDO and cold ENSO events (around 2000, 1956, and 1894), the residua showed a relatively negative value, indicative of lower phytoplankton biomass (Figure 6). During the positive phase of PDO, a southerly anomalous wind off the Philippines is formed, which leads to a northward shift of the North Equatorial Current bifurcation latitude compared with a normal year (Chang and Oey, 2012; Wu, 2013; Figure 7b). Also, more abundant westward-propagating mesoscale eddies frequently approach east of Taiwan and strike the Kuroshio off east Taiwan in 20°–23° N region, which result in greater Kuroshio transport off northeast Taiwan into ECS (Chang and Oey, 2012; Hwang et al., 2004; Wu, 2013; Figure 7b). The occurrence and/or strengthening of a warm ENSO event in the low latitude equatorial area would also cause anomalous westerly winds in the equatorial area (5° S–10° N), strengthen the North Equatorial Current, and enhance the net northward transport of Kuroshio Warm Current (KWC) across 18° N (Kashino et al., 2009; Kim et al., 2004; Yang et al., 2014; Figure 7c). Although the occurrence of ENSO events would cause a shift of North Equatorial Current bifurcation latitude (Kim et al., 2004), its influence is limited and governed by PDO to some degree (Wu, 2013). Furthermore, monsoonal winds in the Changjiang watershed would also be influenced by the PDO and ENSO, which would result in variations in the precipitation and Changjiang water discharge. For example, larger amounts of precipitation and higher Changjiang runoff usually occur during the positive phase of PDO and/or warm ENSO events (Jiang et al., 2006; Xue and Liu, 2008; Wang et al., 2005; Wu, 2013). Higher Changjiang runoff would result in enhanced offshore and alongshore transport of low-density CDW, and lead to an intensified shoreward transport and upwelling of KSW in the Zhe-Min Coast due to the buoyancy effect (Chen, 2008; Wei et al., 2014). Interestingly, chloropigment residua do appear to follow these negative phase of PDO and cold ENSO events (Figure 6), indicating past linkages with phytoplankton biomass.

Figure 6.

Historical variations of sum of residua of all chloropigments, PDO index, and ENSO index. The PDO index data were collected from Shen et al. (2006) and Wu (2013). Data about ENSO index were collected from Yin et al. (2009).

Figure 7.

The schematic summaries of the regional oceanic responses to climate oscillations during (a) normal year, (b) positive phase of PDO, and (c) warm ENSO event (according to Chang and Oey, 2012; Kashino et al., 2009; Kim et al., 2004; Lukas et al., 1991; Wu, 2013; Yang et al., 2014). Arrows indicate the direction of the currents and winds. Orange circles represent eddies.

The statistical analysis between residua value and strength of climate phenomena during the same year or adjacent years (±5 year) showed that there were nearly no correlation between them (R_PDO = 0.25, p = 0.11, n = 41; R_ENSO = 0.02, p = 0.91, n = 32). As previously discussed, many factors could result in this decoupling between residua value and strength of climate phenomena, such as the timeframe difference of our data and the climate data, sectioning precision, and possible errors of core chronology. Other factors such as nutrient inputs, sea surface temperature, and illumination intensity would affect the growth of phytoplankton, and the subsequent deposition and preservation of phytoplankton debris in the sediments would also affect the coupling between phytoplankton biomass in the water column and sedimentary phytoplankton pigment records. These complicated mechanisms and processes would probably enhance decoupling and weaken the correlation between climate phenomena (i.e. PDO and ENSO events) and the response of phytoplankton. The correlations between residua and PDO and/or ENSO events were weaker than between lignin content and strength of EAWM. This was probably due to a more direct physical forcing from the southward delivery of terrestrial materials by the southward flowing coastal current dominated by EAWM.

Previous studies have indicated that the Changjiang water discharge would increase during positive phase of PDO and warm ENSO events (Jiang et al., 2006; Wang et al., 2005; Wu, 2013; Xue and Liu, 2008), and that the CDW is critical in affecting the dynamics of surface waters of Changjiang estuarine area (e.g. low salinity and density and high DIN/DIP ratios; Chen et al., 2004; Shi et al., 2013). Under these circumstances, the influence of the KSW on the phytoplankton biomass in the Changjiang estuarine area may be limited, even during positive phase of PDO and warm ENSO events. Moreover, although increasing Changjiang water discharge would enhance the upwelling of KSW in the Zhe-Min Coast to a certain degree (Chen, 2008; Wei et al., 2014), its influence on the upwelling and phytoplankton growth in this region was probably limited and weaker than the effects of climate variability. Previous studies have also indicated that upwelling in the Zhe-Min coastal area was mainly caused by the orographic uplift and offshore wind (Chen et al., 2004; Shi et al., 2013).

Conclusion

Intensified human activities in the Changjiang drainage basin (e.g. land-use changes, deforestation, and dam construction) post-1970s have enhanced the channel erosion of deeper soils from the watershed to the coast and changed the character of terrestrial OC entering the ECS inner shelf. This has further reduced the lignin content and increased the diagenetic state of terrestrial OC. Additionally, flooding events in the Changjiang basin, hydrodynamic forcing, and regional climate variability (through monsoonal circulation) also significantly influenced the terrestrial OC signature that is delivered to the ECS inner shelf. The low Λ₈ of OC in the Zhe-Min coastal sediments was in part due to the occurrence of the worst flooding events in the Changjiang watershed, which was different from inputs of terrestrial lignin (high Λ₈) to Changjiang estuarine sediments. This difference was likely due to the selective transport of terrestrial lignin detrital materials by hydrodynamic sorting processes. The higher Λ₈ in the Zhe-Min Coast was more correlated to the strengthening of EAWM, which enhanced the southward delivery of lignin-rich materials from Changjiang estuarine sediments.

The intrusion of phosphate-rich KSW and TWWC to the ECS inner shelf is largely controlled by the regional climate variability (i.e. PDO and ENSO), and also had significant effects on the evolution of phytoplankton biomass on the Zhe-Min Coast. The high phytoplankton biomass was likely linked with enhanced intrusion of KSW and TWWC into the ECS inner shelf during the positive phase of PDO and/or warm ENSO events. In contrast, low phytoplankton biomass was more linked to the negative phase of PDO and/or cold ENSO events, during which the intrusion of KSW and TWWC into the ECS inner shelf was relatively weak. The occurrence of HABs and/or increase of phytoplankton biomass in the Changjiang Estuary during past decades may be attributed to the increasing anthropogenic activities in the watershed, but changes in the occurrence of HABs in the Zhe-Min Coast were likely from the intrusion of phosphate-rich KSW that can be linked with regional climate variability.

Footnotes

Acknowledgements

We thank the crews of R/V Runjiang 1 for sampling assistance. We also thank the two anonymous reviewers for all valuable comments on a previous draft. This is MCTL contribution No. 57. Supplementary material to this paper can be found online at .

Funding

This study was supported by the National Natural Science Foundation of China (41176063 and 41221004), and the 111 project (B13030).

References

Andersson

(2011) A systematic examination of a random sampling strategy for source apportionment calculations. Science of the Total Environment 412–413: 232–238.

Barange

Perry

(2009) Physical and ecological impacts of climate change relevant to marine and inland capture fisheries and aquaculture. In: Cochrane K, De Young C, Soto D et al. (eds) Climate Change Implications for Fisheries and Aquaculture: Overview of Current Scientific Knowledge (FAO Fisheries and Aquaculture Technical Paper no. 530). Rome: FAO, pp. 7–106.

Bianchi

Allison

(2009) Large-river delta-front estuaries as natural ‘recorders’ of global environmental change. Proceedings of the National Academy of Sciences of the United States of America 106: 8085–8092.

Bianchi

Mitra

McKee

(2002) Sources of terrestrially-derived organic carbon in lower Mississippi River and Louisiana shelf sediments: Implications for differential sedimentation and transport at the coastal margin. Marine Chemistry 77: 211–223.

Blair

Aller

(2012) The fate of terrestrial organic carbon in the marine environment. Annual Review of Marine Science 4: 401–423.

Caballero-Alfonso

Carstensen

Conley

(2015) Biogeochemical and environmental drivers of coastal hypoxia. Journal of Marine Systems 141: 190–199.

Carpenter

(2005) Eutrophication of aquatic ecosystems: Bistability and soil phosphorus. Proceedings of the National Academy of Sciences of the United States of America 102: 10002–10005.

Carreira

Wagener

ALR

Readman

. (2002) Changes in the sedimentary organic carbon pool of a fertilized tropical estuary, Guanabara Bay, Brazil: An elemental, isotopic and molecular marker approach. Marine Chemistry 79: 207–227.

Chang

Oey

(2012) The Philippines–Taiwan oscillation: Monsoonlike interannual oscillation of the subtropical–Tropical western north pacific wind system and its impact on the ocean. Journal of Climate 25: 1597–1618.

10.

Chavez

Messié

Pennington

(2010) Marine primary production in relation to climate variability and change. Annual Review of Marine Science 3: 227–260.

11.

Chen

(2008) Distributions of nutrients in the East China Sea and the South China Sea connection. Journal of Oceanography 64: 737–751.

12.

Chen

Sheu

(2006) Does the Taiwan Warm Current originate in the Taiwan Strait in wintertime? Journal of Geophysical Research 111: 153–172.

13.

Chen

Bianchi

McKee

(2005) Early diagenesis of chloropigment biomarkers in the lower Mississippi River and Louisiana shelf: Implications for carbon cycling in a river-dominated margin. Marine Chemistry 93: 159–177.

14.

Chen

Bianchi

Mckee

. (2001) Historical trends of hypoxia on the Louisiana shelf: Application of pigments as biomarkers. Organic Geochemistry 32: 543–561.

15.

Chen

YLL

Chen

Gong

. (2004) Phytoplankton production during a summer coastal upwelling in the East China Sea. Continental Shelf Research 24: 1321–1338.

16.

Chu

Zhai

Zhang

. (2007) Effects of Three Gorges Reservoir water storages in 2003 on the median grain size of suspended sediment of the Yangtze River. Transactions of Oceanology and Limnology 3: 23–28 (in Chinese with English abstract).

17.

Conley

Paerl

Howarth

. (2009) Controlling eutrophication: Nitrogen and phosphorus. Science 323: 1014–1015.

18.

Dai

Zhang

. (2011) Variation of riverine material loads and environmental consequences on the Changjiang (Yangtze) estuary in recent decades (1955–2008). Environmental Science and Technology 45: 223–227.

19.

Duan

Wang

. (2005) ²¹⁰Pb distribution of the Changjiang estuarine sediment and the implications to sedimentary environment. Acta Sedimentologica Sinica 23: 514–522 (in Chinese with English abstract).

20.

Duan

Bianchi

(2006) Seasonal changes in the abundance and composition of plant pigments in particulate organic carbon in the lower Mississippi and Pearl Rivers. Estuaries and Coasts 29: 427–442.

21.

D’Arrigo

Wilson

Panagiotopoulos

. (2005) On the long-term interannual variability of the east Asian winter monsoon. Geophysical Research Letters 32: 97–116.

22.

Eadie

Mckee

Lansing

. (1994) Records of nutrient-enhanced coastal ocean productivity in sediments from the Louisiana continental-shelf. Estuaries 17: 754–765.

23.

Fujii

Murashige

Ohnishi

. (2002) Decomposition of phytoplankton in seawater. Part I: Kinetic analysis of the effect of organic matter concentration. Journal of Oceanography 58: 433–438.

24.

Gao

(2002) Shallow marine sedimentation and physical environment evolution as a part of global change: An example from the Bohai, Yellow and East China Sea regions. Earth Science Frontiers 9: 329–335.

25.

Gong

Chang

Chiang

. (2006) Reduction of primary production and changing of nutrient ratio in the East China Sea: Effect of the Three Gorges Dam. Geophysical Research Letters 33: 526–530.

26.

Goñi

Hedges

(1995) Sources and reactivities of marine-derived organic matter in coastal sediments as determined by alkaline CuO oxidation. Geochimica et Cosmochimica Acta 59: 2965–2981.

27.

Gregg

Conkright

Ginoux

. (2003) Ocean primary production and climate: Global decadal changes. Geophysical Research Letters 30: 157–168.

28.

Guo

Yang

Zhang

. (2002) Seasonal distribution of suspended matter in the northern East China Sea and barrier effect of current circulation on its transport. Acta Oceanologica Sinica 24: 71–80 (in Chinese with English abstract).

29.

Chen

Yao

. (2009) Behavior of different phosphorus species in suspended particulate matter in the Changjiang estuary. Chinese Journal of Oceanology and Limnology 27: 859–868.

30.

Hedges

Ertel

(1982) Characterization of lignin by gas capillary chromatography of cupric oxide oxidation products. Analytical Chemistry 54: 174–178.

31.

Hedges

Mann

(1979) The characterization of plant tissues by their lignin oxidation products. Geochimica et Cosmochimica Acta 43: 1803–1807.

32.

Heim

Schmidt

MWI

(2007) Lignin turnover in arable soil and grassland analysed with two different labelling approaches. European Journal of Soil Science 58(3): 599–608.

33.

Houel

Louchouarn

Lucotte

. (2006) Translocation of soil organic matter following reservoir impoundment in boreal systems: Implications for in situ productivity. Limnology and Oceanography 51: 1497–1513.

34.

Zhao

. (2014) Late Holocene elemental and isotopic carbon and nitrogen records from the East China Sea inner shelf: Implications for monsoon and upwelling. Marine Chemistry 162: 60–70.

35.

Sun

Peng

. (2009a) Spatial and temporal variation of organic carbon in the northern South China Sea revealed by sedimentary records. Quaternary International 206: 46–51.

36.

Guo

Feng

. (2009b) Distributions and sources of bulk organic matter and aliphatic hydrocarbons in surface sediments of the Bohai Sea, China. Marine Chemistry 113: 197–211.

37.

Huh

(1999) Sedimentation dynamics in the East China Sea elucidated from ²¹⁰Pb, ¹³⁷Cs, ^{239, 240}Pu. Marine Geology 160: 183–196.

38.

Hwang

Kao

. (2004) TOPEX/Poseidon observations of mesoscale eddies over the Subtropical Countercurrent: Kinematic characteristics of an anticyclonic eddy and a cyclonic eddy. Journal of Geophysical Research 109: C08013.

39.

Jiang

Zhang

Zhu

. (2006) Yangtze floods and droughts (China) and teleconnections with ENSO activities (1470–2003). Quaternary International 144: 29–37.

40.

Jiang

Liu

Chen

. (2014) Responses of summer phytoplankton community to drastic environmental changes in the Changjiang (Yangtze River) estuary during the past 50 years. Water Research 54: 1–11.

41.

Kashino

Espana

Syamsudin

. (2009) Observations of the North Equatorial Current, Mindanao Current, and Kuroshio current system during the 2006/07 El Niño and 2007/08 La Niña. Journal of Oceanography 65: 325–333.

42.

Kim

Jensen

. (2004) Seasonal and interannual variations of the North Equatorial Current bifurcation latitude in a high-resolution OGCM. Journal of Geophysical Research 109: C03040.

43.

Komada

Anderson

Dorfmeier

. (2008) Carbonate removal from coastal sediments for the determination of organic carbon and its isotopic signatures, δ¹³C and Δ¹⁴C: Comparison of fumigation and direct acidification by hydrochloric acid. Limnology and Oceanography: Methods 6: 254–262.

44.

Lewitus

Horner

Caron

. (2012) Harmful algal blooms along the North American west coast region: History, trends, causes, and impacts. Harmful Algae 19: 133–159.

45.

Yao

Bianchi

. (2014a) Organic carbon cycling in sediments of the Changjiang Estuary and adjacent shelf: Implication for the influence of Three Gorges Dam. Journal of Marine Systems 139: 409–419.

46.

Tang

Shi

. (2014b) Increased nutrient loads from the Changjiang (Yangtze) River have led to increased Harmful Algal Blooms. Harmful Algae 39: 92–101.

47.

Bianchi

Allison

. (2012) Composition, abundance and age of total organic carbon in surface sediments from the inner shelf of the East China Sea. Marine Chemistry 145–147: 37–52.

48.

Bianchi

Allison

. (2013) Historical reconstruction of organic carbon decay and preservation in the sediment on the East China Sea shelf. Journal of Geophysical Research: Biogeosciences 118: 1079–1093.

49.

Bianchi

Yang

. (2011) Historical trends of hypoxia in Changjiang River estuary: Applications of chemical biomarkers and microfossils. Journal of Marine Systems 86: 57–68.

50.

Liu

. (2006) Sedimentary features of the Yangtze River-derived along-shelf clinoform deposit in the East China Sea. Continental Shelf Research 26: 2141–2156.

51.

Liu

. (2007) Flux and fate of Yangtze River sediment delivered to the East China Sea. Geomorphology 85: 208–224.

52.

Liu

Yan

Lee

. (2015) Impacts of increasing dissolved inorganic nitrogen discharged from Changjiang on primary production and seafloor oxygen demand in the East China Sea from 1970 to 2002. Journal of Marine Systems 141: 200–217.

53.

Liu

Shi

Liu

. (2010) Records of the East Asian winter monsoon from the mud area on the inner shelf of the East China Sea since the mid-Holocene. Chinese Science Bulletin 55: 2306–2314.

54.

Louchouarn

Lucotte

Farella

(1999) Historical and geographical variations of sources and transport of terrigenous organic matter within a large-scale coastal environment. Organic Geochemistry 30: 675–699.

55.

Lukas

Firing

Hacker

. (1991) Observations of the Mindanao current during the Western Equatorial Pacific Ocean Circulation Study. Journal of Geophysical Research 96: 7089–7104.

56.

Luo

Yang

Zhang

. (2012) Recent spatial pattern and temporal variation in sediment grain size in the inshore area adjacent to the Yangtze Estuary and Hangzhou Bay. Acta Sedimentologica Sinica 30: 137–147.

57.

Mayer

Schick

Allison

. (2007) Marine vs. terrigenous organic matter in Louisiana coastal sediments: The uses of bromine: Organic carbon ratios. Marine Chemistry 107: 244–254.

58.

Meng

Yao

Bianchi

. (2015) Detrital phosphorus as a proxy of flooding events in the Changjiang River Basin. Science of the Total Environment 517: 22–30.

59.

Milliman

Farnsworth

(2011) River Discharge to the Coastal Ocean: A Global Synthesis. New York: Cambridge University Press.

60.

Rong

(2012) Tidal effects on the bulge region of Changjiang River plume. Estuarine, Coastal and Shelf Science 97: 149–160.

61.

Savage

Leavitt

Elmgren

(2010) Effects of land use, urbanization, and climate variability on coastal eutrophication in the Baltic Sea. Limnology and Oceanography 53: 1033–1046.

62.

Schüller

Bianchi

. (2014) Historical reconstruction of phytoplankton composition in estuaries of Fiordland, New Zealand: The application of plant pigment biomarkers. Estuaries and Coasts 38: 56–71.

63.

Shen

Wang

Gong

. (2006) A Pacific Decadal Oscillation record since 1470 AD reconstructed from proxy data of summer rainfall over eastern China. Geophysical Research Letter 33: L03702.

64.

Shi

Wang

. (2013) Taiwan warm current and its impact on the areas of frequent harmful algal bloom in the East China Sea in summer. Oceanologia Et Limnologia Sinica 44: 1208–1215 (in Chinese with English abstract).

65.

Shi

Jiang

. (2004) Preliminary analysis on the relation between the evolution of heavy floods in the Yangtze River catchment and the climate changes since 1840. Journal of Lake Sciences 16: 290–297.

66.

Stephens

Kadko

Smith

. (1997) Chlorophyll-a and pheopigments as tracers of labile organic carbon at the central equatorial Pacific seafloor. Geochimica et Cosmochimica Acta 61: 4605–4619.

67.

Sun

Aller

Lee

(1991) Early diagenesis of chlorophyll-a in Long Island Sound sediments: A measure of carbon flux and particle reworking. Journal of Marine Research 49: 379–401.

68.

Tian

Saliot

(1993) Biogeochemical processes controlling nutrients at the turbidity maximum and the plume water fronts in the Changjiang Estuary. Biogeochemistry 19: 83–102.

69.

Wang

Wei

Chen

. (2012) Annual cycle of hypoxia off the Changjiang (Yangtze River) Estuary. Marine Environmental Research 77: 1–5.

70.

Wang

Yao

Bianchi

. (2015) The effect of particle density on the sources, distribution, and degradation of sedimentary organic carbon in the Changjiang Estuary and adjacent shelf. Chemical Geology 402: 52–67.

71.

Wang

Zheng

Tada

. (2014a) Millennial-scale East Asian Summer Monsoon variability recorded in grain size and provenance of mud belt sediments on the inner shelf of the East China Sea during mid-to late Holocene. Quaternary International 349: 79–89.

72.

Wang

Gao

. (2014b) Sediment records of environmental changes in the south end of the Zhejiang-Fujian coastal mud area during the past 100 years. Chinese Journal of Oceanology and Limnology 32: 899–908.

73.

Wang

Kiyotoshi

Rong

(2005) Decadal climate variability of rainfall around the middle and lower reaches of the Yangtze River and atmospheric circulation. Journal of Tropical Meteorology 11: 169–177.

74.

Wang

Chen

. (2007) Plume front and suspended sediment dispersal off the Yangtze (Changjiang) River mouth, China during non-flood season. Estuarine, Coastal and Shelf Science 71: 60–67.

75.

Ward

Keil

Medeiros

. (2013) Degradation of terrestrially derived macromolecules in the Amazon River. Nature Geoscience 6: 530–533.

76.

Waterson

Canuel

(2008) Sources of sedimentary organic matter in the Mississippi River and adjacent Gulf of Mexico as revealed by lipid biomarker and δ¹³C_TOC analyses. Organic Geochemistry 39: 422–439.

77.

Wei

Chang

Dai

(2014) Streamflow changes of the Changjiang (Yangtze) River in the recent 60 years: Impacts of the East China summer monsoon, ENSO, and human activities. Quaternary International 336: 98–107.

78.

Wright

Jeffrey

Mantoura

RFC

. (1991) Improved HPLC method for the analysis of chlorophylls and carotenoids from marine phytoplankton. Marine Ecology Progress Series 77: 183–196.

79.

(2013) Interannual modulation of the Pacific Decadal Oscillation (PDO) on the low-latitude western North Pacific. Progress in Oceanography 110: 49–58.

80.

Xia

Xie

. (1999) ¹³⁷Cs and ²¹⁰Pb profiles of the seabed cores along the East China Sea coast and their implications to sedimentary environment. Donghai Marine Science 17: 20–27 (in Chinese with English abstract).

81.

Bianchi

Allison

. (2014) Using multi-radiotracer technique to evaluate sedimentary dynamics of reworked muds in the Changjiang River Estuary and East China Sea. Marine Geology, submitted.

82.

Yang

Zhan

. (2011) Influence of the impoundment of the Three Gorges Reservoir on the flux and isotopic composition of particular organic carbon in the lower Changjiang mainstream. Geochimica 40: 199–208 (in Chinese with English abstract).

83.

Milliman

(2009) Seasonal variations of sediment discharge from the Yangtze River before and after impoundment of the Three Gorges Dam. Geomorphology 104: 276–283.

84.

Xue

Liu

(2008) The influence of moderate ENSO on summer rainfall in eastern China and its comparison with strong ENSO. Chinese Science Bulletin 53: 791–800.

85.

Yang

. (2014) A study of correlation between Pacific Decadal Oscillation and the strength of Kuroshio intrusion into the South China Sea. Acta Oceanologica Sinica 36(7): 17–26 (in Chinese with English abstract).

86.

Yang

Zhang

. (2011a) Burial of terrestrial and marine organic carbon in Jiaozhou Bay: Different responses to urbanization. Regional Environmental Change 11: 707–714.

87.

Yang

Milliman

. (2011b) 50,000 dams later: Erosion of the Yangtze River and its delta. Global and Planetary Change 75: 14–20.

88.

Yang

Zhang

Zhu

. (2005) Impact of dams on Yangtze River sediment supply to the sea and delta intertidal wetland response. Journal of Geophysical Research 110: F03006.

89.

Yao

Zhao

Bianchi

. (2014) Remineralization of sedimentary organic carbon in mud deposits of the Changjiang Estuary and adjacent shelf: Implications for carbon preservation and authigenic mineral formation. Continental Shelf Research 91: 1–11.

90.

Yin

Chen

(2009) Relationship between flood/drought disasters and ENSO from 1857 to 2003 in the Taihu Lake basin, China. Quaternary International 208: 93–101.

91.

(2007) Transport and burial of particulate terrigenous organic matter in Changjiang (Yangtze River) and East China Sea Shelf-illustrated by lignin. Master Thesis, East China Normal University (in Chinese with English abstract).

92.

Zhang

. (2007) The characteristics of lignin of plant and soil samples in the Yangtze River (Changjiang) drainage basin. Acta Scientiae Circumstantiae 27: 817–823 (in Chinese with English abstract).

93.

Song

. (2012) Geochemical records of decadal variations in terrestrial input and recent anthropogenic eutrophication in the Chiangjiang Estuary and its adjacent waters. Applied Geochemistry 27: 1556–1566.

94.

Zhang

Jennerjahn

. (2007) Distribution of organic matter in the Changjiang (Yangtze River) Estuary and their stable carbon and nitrogen isotopic ratios: Implications for source discrimination and sedimentary dynamics. Marine Chemistry 106: 111–126.

95.

Zhang

Xue

Wang

. (2014) The spatiotemporal distribution of dissolved inorganic and organic carbon in the main stem of the Changjiang (Yangtze) River and the effect of the Three Gorges Reservoir. Journal of Geophysical Research: Biogeosciences 119: 741–757.

96.

Zhao

Bianchi

. (2012) Historical eutrophication in the Changjiang and Mississippi delta-front estuaries: Stable sedimentary chloropigments as biomarkers. Continental Shelf Research 47: 133–144.

97.

Zhou

Shen

(2008) Responses of a coastal phytoplankton community to increased nutrient input from the Changjiang (Yangtze) River. Continental Shelf Research 28: 1483–1489.

98.

Zhu

Wagner

Talbot

. (2013) Mechanistic controls on diverse fates of terrestrial organic components in the East China Sea. Geochimica et Cosmochimica Acta 117: 129–143.

99.

Zhu

Zhang

. (2014) Reconstruction of anthropogenic eutrophication in the region off the Changjiang Estuary and central Yellow Sea: From decades to centuries. Continental Shelf Research 72: 152–162.

100.

Zhu

Zhang

. (2011) Hypoxia off the Changjiang (Yangtze River) Estuary: Oxygen depletion and organic matter decomposition. Marine Chemistry 125: 108–116.

101.

Zimmerman

Canuel

(2000) A geochemical record of eutrophication and anoxia in Chesapeake Bay sediments: Anthropogenic influence on organic matter composition. Marine Chemistry 69: 117–137.