The Laizhou bay is a semi-closed bay of the Bohai Sea, China, covering about ~6000 km² area with around ~320 km coastline (Wang et al., 2020). The Laizhou Bay is generally shallow, with the average water depth of ~9 m (Jiang et al., 2018). Multiple rivers ( Figure 2 ) discharge terrestrial materials and nutrients to the bay each year (Jiang et al., 2005; Guo et al., 2015; Wang et al., 2019b; Zhang et al., 2019; Peng et al., 2021). The highest nutrient concentrations normally occurred in summer due to the strong river runoff in wet season (Peng et al., 2021). Rivers flowing into Laizhou Bay, like the Xiaoqing River (XQR), were severely contaminated by industrial and agricultural sewage, which could further affect the coastal ecological environment and biogeochemical cycles in Laizhou Bay (He et al., 2018a; He et al., 2018b). These processes and nearshore marine aquaculture activities (Cui et al., 2005) have resulted in low subsurface dissolved oxygen (DO) concentration in the central Bohai Sea during summer (Song et al., 2016; Zhai et al., 2019; Zheng et al., 2020), which also influenced the distribution of hypoxic zone in adjacent regions via hydrodynamic processes (Wei et al., 2019). Hypoxia in sea water has been observed in the Xiaoqing River Estuary, which is strongly impacted by anthropogenic activities (He et al., 2018b). Moreover, the river discharge and the south Bohai anti-clock coastal current results in rather complicated hydrologic conditions for this small bay (Qiao et al., 2010; Zhou et al., 2017; Wei et al., 2019). The eutrophication area and hydrologic condition of Laizhou Bay were closely linked to the river input, which might reveal the potential differences of biogeochemical process in nearshore area and offshore area (Jiang et al., 2018; Zhang et al., 2019; Peng et al., 2021).

We divided our study area into the sub-regions as ‘estuary’, ‘nearshore areas’ closed to river mouths, and ‘offshore areas’ to discuss the source of ladderanes. The estuarine area includes the small river estuaries along the south and east bank of Laizhou Bay. This area was dominated by riverine lateral transportation from upper stream, showing brackish condition between river freshwater discharge and sea water intrusion. The biogeochemical process in these small rivers were less studied comparing with the Yellow River although they were also impacted by strong anthropogenic activity with the development of agriculture and industry in the northern Shandong Peninsula (Zhang et al., 2015; He et al., 2018a). The nearshore ‘area A’ ( Figure 2 ) along the coast was intensely influenced by river input (Zhang et al., 2015; Jiang et al., 2018). The offshore ‘area B’ ( Figure 2 ) was impacted by the Yellow River discharge and Bohai Sea circulations. Although the Yellow River contributed significantly to the hydrology and ecology condition in the northwestern bay (Liu et al., 2012; Peng et al., 2021), anammox activity in the Yellow River Estuary was proved to be very weak comparing with that of the Bohai Sea (Zhang et al., 2017b; Zhang et al., 2018).

Twenty-one suspended particulate samples (6 m depth) from Laizhou Bay were collected in August 2018, and eight suspended particulate samples from estuaries were collected in August 2020 ( Figure 2 ). The sea waters for particulates collection were obtained by submerged pump to large High-Density Polyethylene (HDPE) bottles on board aboard a small vessel of LuChangYu 60003, and about 25 L of water at each station was filtered through pre-combusted glass fiber filters (Whatman GF/F, 150 mm) to get the suspended particulates. The collection of particulate samples from the inner bay in 2018 and from the estuaries in 2020 were both collected in summer, representing similar hydrodynamic conditions in wet season. Specifically, mean water temperature in Laizhou Bay were 27.2 °C and 26.9 °C in the summer of 2018 and 2020, respectively, and water salinity in Laizhou Bay were 30.9 and 30.0 in the summer of 2018 and 2020, respectively (Lu, 2022). Higher temperature and lower salinity both occurred in the western bay in 2018 and 2020 (Lu, 2022). Moreover, according to the monitoring data of water quality, biodiversity and pollution discharge in the Laizhou Bay, the river flows and sea water environments of 2018 and 2020 were similar, e.g., highest DIN concentration both occurred near the XQR estuary at 38.2 and 33.5 μmol/L in 2018 and 2020, respectively (Lu, 2022). We herein suggest our data of particulate samples in estuaries and embayment were comparable.

Water temperature and salinity were measured by a portable multiparameter probe (WTW Multi 3420). Dissolved oxygen (DO) was measured by an automatic potentiometric titrator (Metrohm 848 Titrino plus) after in situ water was additionally collected by Polymethyl Methacrylate (PMMA) hydrophore. Water samples for nutrient analysis were filtrated through GF/F filters (Whatman) and immediately frozen at –20°C. Dissolved inorganic nitrogen (DIN) in seawater including NH 4 + , NO 2 − and NO 3 − were quantified using flow injection analysis (QuAAtro, Seal, Germany) after thawing the samples at 4 °C. Precision of measurements above were 0.1 °C for water temperature, 0.1 for salinity, 0.1 mg/L for DO and 0.01 μmol/L for nutrients. The detection limits were 0.03 μmol/L for NH 4 + , 0.02 μmol/L for NO 2 − , 0.05 μmol/L for NO 3 − . All these parameters were used as the average value after triple measurement.

Measurement of bulk organic carbon in particulate samples followed the procedure of Hu et al. (2016). Suspended particulates on filters were first freeze-dried and then treated with 4 mol/L HCl to remove inorganic carbon. After drying in an oven at 55°C for 24 hours to volatilize the excess HCl, filters were wrapped in tin boat. Particulate organic carbon (POC) was analyzed with a CHN element analyzer (Thermo Flash EA2000). Atropine Standard (OC = 70.50 ± 0.24%) and Low Organic Content Soil Standard OAS (OC = 1.61 ± 0.09%) were used as standard samples, and the relative deviation was less than 5% (n = 6). The POC contents were reported as μg/L in seawater of this study, with the detection limit of 3 μg/L.

Terrestrial and marine biomarkers ratio, abbreviated as TMBR, has been used to evaluate the relative contribution of terrestrial organic carbon to the total organic carbon (Xing et al., 2011). TMBR′ has also been used by using n-alkanols in the calculation to replace the n-alkanes which are susceptible to pollution (Yuan et al., 2013). TMBR′ is calculated using the following equation:

where Σ(C₂₆ + C₂₈ + C₃₀ n-alkanols) represents the terrestrial input. Σ(B + D + A) is the sum of marine biomarker concentrations, referring to brassicasterol (B), dinosterol (D) and C₃₇ alkenones (A), respectively. Alkenones were below detection limit in all samples from Laizhou Bay, thus the sum of marine biomarker concentrations in this study was represented as Σ(B + D).

The NL₅ ratio has been widely used to infer ladderane synthesis temperature. In order to keep the fluidity of anammoxosome membrane in different environment, the membrane lipid chain length of anammox bacteria is temperature dependent. NL₅ is calculated using following equations (Rattray et al., 2010):

Distribution maps of sea water temperature, salinity, DO, DIN, POC, biomarker concentrations and biomarker indices were generated using software Ocean Data View (Schlitzer, 2021). Maps were created using weight-average gridding loaded by Ocean Data View version 5.6.3.

Statistical tests including t test and Pearson’s correlation analysis were performed with OriginPro 2018C software. A probability (p) of 0.05 was used to judge the statistical significance.

The water temperature, salinity and DO were measured both in estuaries and bay. The water temperature in estuaries was in the range of 27.5–29.2°C, slightly higher than that from the area A (26.0–29.2°C) and area B (25.2–29.1°C) ( Figure 3A ). Water temperature in the bay was the highest in the most southwestern site of area A nearby the XQR estuary ( Figure 3A ). The salinity ranged from 2.0 to 12.0 in waters from the estuaries ( Figure 3B ), showing a brackish condition between river freshwater discharge and sea water intrusion. In the aera A and area B, the salinity varied from 28.4–31.8 and 29.3–32.2, respectively, with lower salinity in the western bay ( Figure 3B ). DO concentrations in estuaries varied from 1.8 to 8.4 mg/L, with the lowest DO concentration in the XQR Estuary (1.8 mg/L) located at the southwest bank ( Figure 3C ). In the area A, the lowest DO concentration (5.3 mg/L) was found in the southwestern bay nearby the XQR estuary, while the highest value (6.3 mg/L) occurred in the southwestern bay nearby the Nanyang River (NYR) estuary and Jiaolai River (JLR) estuary. In the area B, the lowest DO concentration (4.1 mg/L) was found in the northwestern bay in the adjacent area of the Yellow River Estuary, while the highest value (7.0 mg/L) occurred in the most northeast site ( Figure 3C ). DIN was measured in the bottom water in Laizhou Bay. Generally, Total DIN and individual DIN ( NH 4 + , NO 2 − and NO 3 − ) concentrations were generally higher in the western bay of both area A and B, especially in the southwestern sites of area A nearby the XQR Estuary ( Figure S2 ).

POC concentrations varied from 897 to 4279 μg/L in estuaries, which were much higher than those from the area A (260–767 μg/L) and area B (148–560 μg/L). Maximum POC concentrations were found in the JLR Estuary located at the southeast bank of Laizhou Bay ( Figures 4A , S3A ). In Laizhou Bay, the highest POC concentrations of 767 μg/L and 638 μg/L were detected at two nearshore sites in the area A ( Figure 4A ).

Concentrations of terrestrial biomarker long-chain (C₂₆+C₂₈+C₃₀) n-alkanols were 80–930 ng/L in estuaries, 12–80 ng/L in the area A and 20–180 ng/L in the area B ( Figures 4B , S3B ). Concentrations of marine biomarker sterols (sum of brassicasterol and dinosterol) were 215–1500 ng/L in estuaries, 74–412 ng/L in the area A and 155–409 ng/L in the area B ( Figures 4C , S3C ). Long-chain n-alkanols concentrations were generally high in the western bay ( Figure 4B ), while marine sterols concentrations displayed relatively high values in the central bay ( Figure 4C ). Terrestrial and marine biomarker ratio TMBR′ values were higher in estuaries (0.19–0.61, average 0.33) than that in the area A (0.06–0.41, average 0.20) and area B (0.05–0.41, average 0.17) ( Figure 4D ), which displayed similar distribution pattern with that of n-alkanols ( Figures 4B–D ). Although average TMBR′ values were close between area A (0.20) and area B (0.17), OC sources were quite different. For terrestrial OC, area A was mainly influenced by small river inputs, while area B influenced by Yellow River input ( Figure 4B ). Besides, higher primary productivity as indicated by high sterols concentrations ( Figure 4C ) in offshore area B, resulting in the relative low TMBR′ values in the central and northern of Laizhou Bay.

Anammox biomarker ladderane lipids were detected in all suspended particulates from the estuaries and bay. In estuaries, total ladderane concentrations varied from 1.1 to 6.0 μg/L, with ladderane III and ladderane IV as the two dominating compositions (accounting for 50% and 45% of total ladderanes, respectively) ( Figure 4E ). For the bay, total ladderane concentrations varied from 0.01 to 2.5 μg/L, showing high values in the area A ( Figure 4F ). Ladderane III and ladderane IV were also the two most abundant ladderane in Laizhou Bay, accounting for 43% and 42% of the total ladderanes, respectively ( Figure S4 ).

Ladderanes in particulates and sediments from the open ocean were commonly considered as in situ production (Jaeschke et al., 2007; Jaeschke et al., 2009; Rush et al., 2012a; Rush et al., 2012b; Zhao et al., 2019), however, ladderane sources in coastal seas are more complicated due to riverine inputs and anthropogenic influences (Hou et al., 2013; Hu et al., 2016; Zhang et al., 2017b). In this study, we interpret the distinct differences of ladderane lipid compositions and concentrations from the estuaries (average 2.6 μg/L) to area A (average 0.9 μg/L) and then to area B (average 0.1 μg/L) as broadly reflecting ladderane sources. However, the biosynthetic origins of ladderanes from terrestrial and marine anammox bacteria were not well constrained, which limited our understanding of the controlling mechanism of ladderane sources in coastal area. We herein used a ladderane compound ratio in particulates, and the correlation between ladderanes and terrestrial-marine biomarkers (e.g., Hu et al., 2016; Zhao et al., 2019) to semi-quantitatively estimate the origins of ladderanes in Laizhou Bay. The proportion of individual ladderane compounds varied in different anammox bacteria genera (Rattray et al., 2008). The ladderane components were broadly similar in two terrestrial anammox bacteria (Ca. Brocadia and Ca. Kuenenia) while the marine anammox bacteria Ca. Scalindua showed a notably different ladderane proportion (Rattray et al., 2008). The absolute values of ladderane concentrations from different publications were difficult to be compared due to analytical different methods used. However, the relative proportion of each individual compounds in different environments were comparable according to the Rattray et al. (2010), in which another ladderane ratio index ‘NL₅’ was proposed and been widely used in the following studies. We herein proposed a new index in terms of ladderane ratios, Rlad, to distinguish the biosynthetic origin of ladderanes:

R_lad was the ratio of ladderane lipids (ladderane III: C₂₀-[5]-ladderane FAME and ladderane IV: C₂₀-[3]-ladderane FAME), both occurred widely with higher concentration than lad I and lad II in natural environments (e.g. Hu et al., 2016; Zhao et al, 2019). In a few culture experiments, R_lad value was 1.1 ± 0.2 in terrestrial anammox bacteria Ca. Brocadia and Ca. Kuenenia, and 0.6 in marine anammox bacteria Ca. Scalindua (single measurement) (Rattray et al., 2008). By summarizing published ladderane data (41 samples) of both particulates and sediments, we found R_lad values varied significantly in different environments. R_lad values were generally above 0.9 in coastal areas, for example 1.0–1.8 (average 1.4) in the Senegal River Estuary (Jaeschke et al., 2010), 0.9–9.0 (average 2.3) in the Pearl River Estuary (Zhu, 2014) and 0.9–2.1 (average 1.5) in the Changjiang River Estuary (Hu et al., 2016). On the other, R_lad values were lower than 0.9 in more open ocean environments, such as 0.6–0.8 (average 0.7) in the Celtic Sea (Rattray, 2008; Jaeschke et al., 2009), 0.2–0.3 (average 0.3) in the Arabian Sea (Jaeschke et al., 2007; Rattray, 2008) and 0.1–0.4 (average 0.3) in the Black Sea (Kuypers et al., 2003; Rattray, 2008). Though the absolute ladderane contents might be different in these studies due to the different methods used, the ratio of individual ladderanes were more comparable in the application of R_lad ratio. Thus, we propose that the R_lad values could semi-quantitatively distinguish ladderane from terrestrial or marine anammox bacteria, with R_lad values higher than 0.9 indicating terrestrial source and R_lad values lower than 0.9 indicating marine source.

Our study of the sources and spatial distributions of ladderanes in Laizhou Bay region provides an example of using ladderanes as an indicator to evaluate human influences on anammox activities in the river-coastal water environments. Laizhou Bay has been considered as a typical eutrophic bay since the 1980s (Jiang et al., 2005), with DIN concentration increased of 10 folds from the 1980s to the 2000s followed by gradual reduction since 2010 (Zhu et al., 2019). Nutrients and OC into Laizhou Bay were transported by rivers mostly flow through the Shandong Province, where the population and economy had been growing fast in North China (National Bureau of Statistics of China, 1999–2020). According to the official statistical results of pollution source published by the Shandong Provincial Department of Ecological Environment (http://117.73.254.13:8801/wryfb/MapMainT.html), ~ 100 key sewage outlets and ~ 50 key sewage plants existed in the drainage area of rivers entering Laizhou Bay. Most of these outlets were located in the Xiaoqing River basin and the Jiaolai River basin ( Figure 7 ). The Xiaoqing River, for example, was severely contaminated by anthropogenic discharge especially in wet season. A recent study reported that the Xiaoqing River displayed high discharge flux to Laizhou Bay with 2649.2 t/a for DIN, 89.94 t/a for phosphate and 4495.6 t/a for COD (chemical oxygen demand) (Lu, 2022). Besides, multiple wastewater outlets and sewage treatment plants were located along the river as reported by Shandong Provincial Department of Ecological Environment ( Figure 7 ). The eutrophication and organic matter pollution triggered oxygen depletion along the river and in the estuary (Meng et al., 2005; He et al., 2018a; He et al., 2018b), providing favorable conditions for anammox growth (e.g., Dang et al., 2010; Zhao et al., 2019; Li et al., 2021). The point sources of sewage discharge could be also an important origin of ladderanes in estuaries because anammox bacteria and ladderanes were generally detected in sewage water/sludge (e.g., Ren et al., 2022) and nearby the wastewater outlets (e.g., Shi et al., 2021). Therefore, the ladderane contribution to the river estuaries was closely linked to the contaminated condition of rivers under anthropogenic influences. Inside Laizhou Bay, the distribution of ladderanes was less influenced by the in situ water condition but rather being mostly affected by river inputs or by bay circulation transport.

In order to further evaluate the anammox activity affected by the terrestrial and anthropogenic influences in a wider context, we compared the ladderane concentrations in particulates from Laizhou Bay with published data ( Figure 8 ). Since that ladderane concentrations could reflected the intense of anammox activity (e.g., Pitcher et al., 2011; Rush et al., 2012a; Zhao et al., 2019), we could compare the anammox activity under different source mechanisms in different sea area. It should be noted that the previous studies only focused on the anammox process in sediments of Chinese marginal seas by using molecular techniques and isotope labelling technique (e.g., Dang et al., 2010; Hou et al., 2013; Zhang et al., 2019; Teng et al., 2022), but did not include the potential anammox activity in water particulates. We only compared ladderane concentrations from the China marginal seas because these data were analyzed using same procedures including ladderane extraction, purification, detection and quantification (Zhu, 2014; Hu, 2016; Hu et al., 2016; Shi et al., 2021). Generally, ladderane concentrations at hypoxia sites were higher than other sites in the same study area ( Figure 8 ), which further proved that oxygen level (human influences) was the main controlling mechanism of anammox activity. By comparing ladderane data between different studies, the highest ladderane concentrations occurred in the estuaries of Laizhou Bay (average 2.6 μg/L, this study, Figure 8 ), where ladderanes mainly originated from terrestrial environment under severe anthropogenic influence. Highest ladderane concentrations in Laizhou Bay estuaries indicated the more intense anammox activity in small river estuaries comparing with other coastal areas. Ladderane concentrations in Jiaozhou Bay were also high (average 1.5 μg/L) though no obvious hypoxia occurred (Shi et al., 2021). As a semi-closed small bay in North China, Jiaozhou Bay was also strongly affected by anthropogenic discharge. Ladderanes distribution in Jiaozhou Bay was affected by river flow and wastewater discharge and their concentrations could reflect the bottom water quality condition (Shi et al., 2021). However, comparing with these small-scale systems above, ladderane concentrations were low in large rivers such as the Changjiang River and the Pearl River ( Figure 8 ), which are two of the biggest rivers in China with large terrestrial input (Gao et al., 2002; Sun et al., 2020; Zhou et al., 2022) and occasional hypoxia events in estuaries (Dai et al., 2006; Li et al., 2020). Moreover, anammox activities had been observed in both two rivers and provided evidence for the terrestrial input of ladderanes in estuaries (Hou et al., 2018; Wang et al., 2019a), we thus deduced that the concentrations of terrestrial ladderanes might be diluted by the huge riverine flow. As to the East China Sea (ECS) adjacent to the Changjiang Estuary, ladderanes in the bottom water were proved to be produced in situ rather than terrestrial input (Hu et al., 2016), generally lower than in bays and most estuaries ( Figure 8 ). However, ladderane concentrations in the hypoxic area of the ECS were much higher than other sites of the ECS and comparable with the hypoxic area of the Changjiang Estuary ( Figure 8 ). The bottom hypoxia in the ECS has been occurring since the 1950s (Wei et al., 2007; Zhu et al., 2011; Liu et al., 2020). Increasing DIN discharge from the Changjiang River was the main reason of the occurrence and increasing of hypoxia zone in the East China Sea (Zhang et al., 2007; Wang et al., 2016). In conclusion, high ladderane concentrations in particulates were more likely occurred in small bays and hypoxic areas, furthermore indicated that the anammox activity could be enhanced by anthropogenic influence like sewage discharge, eutrophication and hypoxia. Nitrogen loss by anammox process might be underestimated in such small-scale system. In addition, the main terrestrial source of ladderanes in estuaries could prove key evidence for the importance of terrestrial contribution to coastal anammox process. Therefore, by identifying the sources of the ladderanes in human-influenced bays, we can better evaluate the intensity of anammox activity in these small but vital river and estuarine systems and further estimate the anammox contribution in the nitrogen cycle from coastal seas.