Total Maximum Allocated Loads on Stoichiometry of Nitrogen and Identification of Critical Form in Jiaozhou Bay, China

LIN Guohong, SONG Xianli, LU Dongliang, LI Keqiang,LIANG Shengkang, and WANG Xiulin

Total Maximum Allocated Loads on Stoichiometry of Nitrogen and Identification of Critical Form in Jiaozhou Bay, China

LIN Guohong1), 2),#, SONG Xianli3),#, LU Dongliang4), LI Keqiang1), *,LIANG Shengkang1), and WANG Xiulin1)

1) Key Laboratory of Marine Chemistry Theory and Technology (Ocean University of China), Ministry of Education, Qingdao 266100, China 2) College of Material Science and Engineering,Qingdao University, Qingdao 266100, China 3) Marine Biology Institute of Shandong Province, Qingdao 266104, China 4) Guangxi Key Laboratory of Marine Disaster in the Beibu Gulf, Qinzhou University, Qinzhou 535099, China

Total pollutant load control management for total dissolved nitrogen (TDN) is an urgent task required to gain a good water quality status in Jiaozhou Bay (JZB), China. In this paper, the stoichiometry of multiform TDN on land-ocean interactions associated with marine biogeochemical reaction (LOIMBR) was studied by modeling the load-response relationship based on a three-dimensional water quality model of nitrogen in JZB. The results showed that the stoichiometry on LOIMBR of dissolved organic nitrogen (DON), NO3-N and NH4-N was 3:1:1, with one-third of the contribution on the concentration of dissolved inorganic nitrogen (DIN) in JZB for the land-based DON loads to DIN loads. Based on the stoichiometric relationship of nitrogen forms, the total maximum allocated load (TMAL) of equivalent TDN (ETDN) was approximately 5300ta−1in JZB, equivalent to the TMAL of 5700, 5800 and 15600ta−1for NH4-N, NO3-N and DON, respectively. According to the loads of ETDN, there were four outfalls overloaded in JZB in 2015, which lie in the head of the bay. In the four overloaded outfalls, besides NO3-N, NH4-N was the critical nitrogen control form for Moshui River, while DON for Dagu River and Haibo River. The results of numerical experiments further showed that JZB will achieve good water quality after 7 years by implementation of the ‘different emission reduction’ based on TMAL of ETDN, which is significantly better than ‘equal percent removal’.

total dissolved nitrogen; water quality; stoichiometry; total maximum allocated load; Jiaozhou Bay

1 Introduction

Since the early 1980s, the amount of land-based pollution discharged into Jiaozhou Bay (JZB) has increased year by year with the rapid development of the social economy in Qingdao(Zhang., 2017). As a result, the concentration of DIN has increased continuously in JZB, and the water quality of JZB has deteriorated (Shen., 2006; Dong., 2010; Liang., 2015; Zhang., 2017). Since 2011, according to the equal percent removal (EPR) method, Qingdao has launched a control program with an annual reduction rate of approximately 3% of single NH4-N form (NH4-NLB) of land-based nitrogen loads (Qingdao Environmental Protection Bureau, 2011). In recent years, marine environment quality monitoring has shown that the water quality has deteriorated slowly, but nitrogen concentration is still high in JZB (Wang, 2017; Zhang, 2017). The area of poor status of water quality (worse than the grade II National Seawater Quality Standard, 14μmolL−1for DIN) was as much as 50%, while, there was still 13% worse status (worse than the grade IV National Seawater Quality Standard, 36 μmolL−1for DIN) in JZB (Qingdao Municipal Ocean and fisheries administration, 2002-2017). This means that the EPR pollutant reduction effect is much lower than expected, with low efficiency of NH4-NLBreduction in JZB (Liu., 2005; Yang, 2014; Lu., 2016; Li., 2018; Lu, 2018). The main forms of land-based TDN have changed gradually with industrial restructuring in the past 20 years, especially the proportion of DON, increasing from approximately 25% in 1995 to the current level of 40% (Zhang, 2013; Liu., 2014), and the proportion of NO3-N, which increased from approximately 16% in 2000 to the current level of 57% (Li., 2018). Thus, finding out the relationship between the different forms of land-based TDN and the concentration of DIN in JZB, and then transforming the EPR to the multiform land- based TDN ‘different emission reduction’ is an urgent task in order to achieve efficient water quality management in JZB.

Previous studies have shown that the effect of land-based pollution ‘different emission reduction’ is obviously better than that of the EPR in the treatment of coastal water quality (Sobel, 1965; USEPA, 2010; Dai., 2015). Thus, in the early 1970s, the total loads control programs in USA, named total maximum daily load (TMDL), widely used the ‘different emission reduction’ (Lung, 2001), which was also practiced by total pollutant load control system (TPLCS) in Japan, and by water framework direct (WFD) and marine strategy framework directive (MSDF) in European Union (Borja, 2005; Hering., 2010). However, they were mainly reflected in the different outlets rather than the different TN forms,.., the TMDL program of TN in Chesapeake watershed since 1992 to reduce hypoxia (Linker., 2013a; Li., 2017). A quantified goal of the Chesapeake Bay Agreement was to implement at least a 40% reduction in the overall nutrient loads entering the main stem of the Bay compared to the benchmark estimated in 1985 (Perciasepe, 1992; Cerco and Cole, 1993; Thoman., 1994). However, with a limited scope of the controllable load definition, the emission reduction actually achieved only a 22% reduction in nitrogen and continued to display poor water quality (Linker., 2013b). In fact, the quality effects on water quality of different land-based TDN forms is needed to accurately identify critical nitrogen forms for TDN total maximum allocated loads (TMAL) in ‘different emission reduction’ management (Dai., 2015).

At present, most studies focus on the TMAL associated with land-based TN effect (Ge., 2003; Han., 2011; Jiang., 2011; Li., 2015). There are no reports about the quantitative relationship between the composition of land-based TDN inputs and the concentration of DIN in sea water. The effect of different land-based TDN input on water quality in coastal water is different, influenced by environmental factors (.., phytoplankton communities) besides the composition of TDN, especially the DON compositions (Beck and Straten, 1983). Research confirms that a large fraction of the DON includes truly recalcitrant forms that persist in the environment for months to hundreds of years (Benner, 2002; Bronk., 2007). While, labile DON, which include urea, dissolve free amino acids and nucleic acid, can turn over to NH4-N photochemically or biologically on the order of seconds to days (Kieber., 1999; Wiegner., 2001; Koopmans., 2002; Bronk, 2007; Simsek., 2013). The dynamics of DON is associated with the sources of DON and environmental conditions, with different residence time of 15 d for DON in Dapeng Bay and 13 d in California Bay (Jorgensen., 1983; Bronk, 2002; Li., 2013). Additionally, the degradation rate of DON is 0.18 to 0.21d−1from rivers but 0.3 to 0.35d−1from sewage treatment plants (Li., 2017). Whilst, Nitrification and denitrification are the most common reactions involving nitrogen in stream systems, and the dominant nitrogen transformation processes (Zarnetske., 2011). Water quality model can be used to quantify the relationship between compositions of land-based TDN inputs and the concentrations of DIN in sea water, and calculate the TMAL of TDN (Srinivasan and Arnold, 1994; Drago., 2001; Howarth., 2002). The precise simulation of spatial and temporal distribution of nitrogen concentrations in coastal sea is primarily to understand the water quality effects of different land-based TDN input (Li., 2015; Lu., 2016). Whilst there have been many water quality-based methods to calculate TMAL since the mid-1960s (Deininger, 1965; Loucks., 1967; Burn and McBean, 1985; Fujiwara., 1986; Ellis, 1987; Boese, 2002; Deng and Zheng., 2010; Han and Li., 2011).

Therefore, the purpose of this paper was focused on reducing the DIN concentration by total pollutant load control management for TDN efficiently in JZB, revealing the water quality effects to identify critical TDN reduction forms. The ‘different emission reduction’ assessment indicators were compiled by calculating the TMALs on stoichiometry of different forms of nitrogen. The results of this research are expected to provide a quantitative scientific basis for the total pollutant load control management for multiform land-based TDN.

2 Methodology

2.1 Study Area

Located in the southern coastal area of the Shandong Peninsula in China and surrounded by the Qingdao City, the Jiaozhou Bay (35˚58´-36˚24´N, 120˚-120˚31´E) is a typical semi-enclosed bay (Fig.1). It has a channel 2.5 km in width connected to the Yellow Sea (Lu., 2016). Its total surface area is 360 km2and the depth is approximately 7-8m (Liu., 2005). The bay is observably affected by anthropogenic activities given the rapid development of surrounding cities. Researches show that the anthropogenic sources (.., sewage, industrial wastes and agricultural runoff) have become the dominate inputs of the nitrogen (Zhang., 2006). A large amount of land-based pollutants entered JZB through seven rivers named Lianwan River, Yang River, Dagu River, Moshui River, Loushan River, Licun River, Haibo River and five wastewater treatment plants (WWTP) named Lianwan WWTP, Loushan WWTP, Licun WWTP, Haibo WWTP and Tuandao WWTP. As the location is adjacent, the estuaries and WWTP should be merged into the 8 outfalls, they are named Lianwan River (LWR), Yang River (YR), Dagu River (DGR), Moshui River (MSR), Loushan River (LSR), Licun River (LCR), Haibo River (HBR) and Tuandao (TD), respectively.

Fig.1 The geographical location of JZB. The blue parts indicate the ocean and rivers. The solid squares, triangles and circles represent the stations, estuaries and WWTPS, respectively, which will be described in the data parts.

2.2 Stoichiometry of Nitrogen on Land-Ocean In-teractions Associated with Marine Biogeo-chemical Reactions

When TDN, including NH4-N, NO3-N, NO2-N and DON, enter into the multimedia marine environment of a bay, they may be subjected to a series of hydrodynamic- biogeochemical processes on transform/transfer land- based TDN to marine DIN. Such as remineralization (mainly for DON and detritus), advection or diffusion, adsorption/desorption of the particles, and biotransformation with phytoplankton uptake, exudation, mortality and zooplankton grazing (Goldman, 1983; Dillon and Molot, 1990; Fasham., 1990; Devol., 1993; Lohse., 1993; Gruber and Sarmiento, 1997; Hydes., 1997; Codispoti., 2001; Fennel., 2006; Yuan., 2012) (Fig.2).

Fig.2 Biogeochemical processes of land-based TDN loads transforming to marine DIN. LB represent the land-based TDN, PPT and ZPT represent phytoplankton and zooplankton, respectively. The numbers on the arrows represent the biogeochemical process. 1 indicates remineralization of land-based TDN; 2 indicates biotransformation; 3 indicates photosynthesis; 4 indicates phytoplankton exudation; 5 indicates zooplankton grazing; 6 indicates phytoplankton mortality; 7 indicates zooplankton mortality; 8 indicates zooplankton exudation; 9 indicates detritus degradation; 10 indicates marine DIN hydrotransformation; 11 indicates detritus subsidence; and 12 indicates detritus suspension.

As for the concentration of DIN (C), it was calculated using the following equation (Li., 2015; Lu., 2017):

whererepresents time;represents the source terms, which are terrestrial input, atmospheric deposition, aquaculture emission,.;biogeochemrepresents the biogeochemical variation terms, which mainly include nitrogen cycle in the marine ecosystems (bio) and nitrogen exchange at the sediment interface (sed):

andhydrorepresents the hydrodynamic exchange terms, which include flowing in and out the bay with advection and diffusion.

Due to its semiclosed physical nature, approximately 80% of the nitrogen pollutants are derived from the land in JZB (Liu., 2005; Wang., 2006; Zhang, 2007). The impact of particulate nitrogen (PN) on DIN concentration is much lower than that of TDN (Song., 2006; Hu., 2009) in JZB, and therefore land-based TDN load control is significantly effective. Thus, if we neglect the other sources, the processes of land-based nitrogen pollutant discharged into a marine multimedia environment can be thought of as a biogeochemical reaction:

where TDNland-basedmeans the land-based loads of nitrogen, including NH4-N, NO3-N, NO2-N and DON; DINinand DINoutrepresents marine dissolved inorganic nitrogen residing in and out the bay by migration and diffusion, including NH4-N, NO3-N and NO2-N; BioCN and SedN represents the organic nitrogen fixed in the organisms and nitrogen fixed in the sediment, with particle nitrogen information, and α, β, γ, δ, and ε represent the stoichiometry relationship.

2.3 Total Maximum Allocated Loads on Stoichiometry

For the 8 outfalls, which are LWR, YR, DGR, MSR, LSR, LCR, HBR and TD, TMAL of TDN can be calculated by the water quality-based simulation-optimization method (Su., 2013; Dai., 2015). While TDN consists of NH4-N, NO3-N, NO2-N and DON, with different biogeochemical processes, the TMAL of TDN should be the sum of TMALs of all nitrogen forms based on stoichiometry. The concentration of DIN in JZB (C) can be expressed as:

where0is the background concentration of DIN, [] is the response concentration matrix of DIN in a bay is the increment of DIN concentration caused by the unit increment on the TDN loads (Lu., 2017) which infers the hydrodynamic-biogeochemical processes on transform/transfer land-based TDN to marine DIN and can be calculated using the three-dimensional water quality model of nitrogen (3DWQMN) in JZB (Lu., 2017). The model adopt an orthogonal system that is spatially curvilinear, with grids that measure 128×92 (rows×co- lumns) and a resolution ranging from a minimum of 284 m in JZB to a maximum of 686m near the boundary of the open ocean (Bao., 1999). Some researches calibrated biogeochemical processes of NH4+, NO3−, NO2−andDON, the dynamic parameters and improved the simulation accuracy (Li., 2015; Lu., 2016). The biogeochemical processes in the three-dimensional water quality model are expressed as follows:

when the water quality criteria (collected from Qingdao Ocean Functional Zoning) are set in JZB, the TMALs of land-based NH4-N, NO3-N, NO2-N and DON can be calculated (Dai., 2015). Thus, the TMAL of ETDN can be expressed as the sum of TMALs of all nitrogen forms based on stoichiometry:

where TMALNis the total maximum allocated loads of theth TDN forms, and αNis the stoichiometry of theth TDN form on land-ocean interactions associated with marine biogeochemical reaction.

2.4 Overload Carrying and Critical Form of Nitrogen Identification

For efficient management to reduce the DIN concentration in JZB, emission reduction is needed for the overload outfalls but exemption for surplus-load and flat-load outfalls. Here, we use the land-based pollutants overload rate to identify the overload, surplus or flat outfalls as follows:

whereis outfalls, TMALETDNis TMAL on stoichiometry, and LoadETDNis the current load on stoichiometry, which can be expressed as the sum of loads of all nitrogen forms (LoadN) based on stoichiometry:

whereis the form of nitrogen.

The over-level-to-area ratio (at the level of grade II National Seawater Quality Standard)and input loads of TDN are used to facilitate comparing the effects of the current EPR method with the land-based TDN ‘different emission reduction’ method.

The over-level-to-area ratio(ES%) is calculated using the following equation:

whereSis the sea area of exceeding National Water Quality grade II, andis the sea area of JZB.

2.5 Data

Field observations were conducted during February (winter), May (spring), August (summer) and November (autumn) 2015 in 20 stations around JZB (Fig.1) (Li., 2018). The main sources of TDN pollutants were land sources, as well as the rivers, ditches, and treated wastewater that enter the bay. Synchronous sampling of land- based nitrogen (NH4-N, NO3-N, NO2-N and DON) inputs was conducted in 11 rivers and WWTPs around JZB. The average monthly land-based TDN loads is 1240 tons and average DIN concentrations is 20.9μmolL−1in JZB. The monitoring results have been shown in Li. (2018). There were seasonal fluctuations of land-based loads of TDN were 1500, 1760, 1150 and 540tm−1in spring, sum- mer, autumn and winter, respectively, with the average percentage of different TDN forms being 47% for DON, 37% for NO3-N, 12% for NH4-N, and 4% for NO2-N. And there were seasonal fluctuations of DIN concentration, changing from 21.3μmolL−1in spring, to 23.3 μmolL−1in summer, 21.3μmolL−1in autumn and 17.7μmolL−1in winter. The These land-sea synchronous monitoring data is used as the input data for the 3DWQMN when calculating the stoichiometric relationship of nitrogen forms and the total maximum allocated loads of TDN.

3 Results

3.1 The Responding Concentrations of DIN

The response concentrations of DIN (response concentration matrix) in JZB were calculated under unit loads of land-based NH4-N, NO3-N and DON, respectively, by using the 3DWQMN model, while the response concentration of NO2-N was neglected for its minor loads (<5%). The DIN concentration of single outfalls is almost fan- shaped as the trend decreases. Superposition of the response concentration matrix is used to integrate the response concentration matrix. Integration results show significant differences in the response concentration of different DIN (NH4-N, NO3-N) and DON loads for DIN (<0.01). The increment of DIN concentrations are approximately 0.034±0.0011μmolL−1(ta−1)−1and 0.031± 0.0014μmolL−1(ta−1)−1for NH4-N and NO3-N loads, respectively, while it is approximately 0.011±0.00050 μmolL−1(ta−1)−1for DON loads, with a ratio of 3:3:1

The distribution patterns of response concentrations of DIN are similar for all land-based TDN loads through four seasons, decreasing from bay-head to bay-mouth in JZB (Fig.3). The DIN concentrations show high values in the northeast corner and western area near the Dagu estuary of JZB. This result might be due to the joint action of land-based TDN inputs and weak hydrodynamic exchanges in the northeast JZB (Liu., 2004). The intensity of response concentrations fluctuate seasonally, decreasing in the sequence of summer > autumn > spring > winter, with average values of approximately 0.034± 0.015μmolL−1(ta−1)−1, 0.028±0.010μmolL−1(ta−1)−1, 0.022±0.013μmolL−1(ta−1)−1and 0.018±0.012μmolL−1(ta−1)−1, respectively.

There is significant difference of response concentrations of DIN to land-based TDNin seasons, which is associated with the transform/transfer process,.., hydrodynamic mixing, biotransformation and natural degradation (Bronk., 1998; Seitzinger., 2001; Liu., 2017). This can be inferred from the land-sea synchronization survey results (Lu., 2016). Research shows that uptake of different nitrogen form by phytoplankton depends on species associated with marine locations and seasons (Berges., 2008; Yin., 2008). Diatoms bias towards NO3-N (Goldman., 1993; Lomas., 2000), while dinoflagellates prefer reduced nitrogen and adapt to the low NO3-N but high NH4-N and urea (Berg., 1997; Carlsoon., 1998). It is generally believed that phytoplankton uptakes NH4-N preferentially, while NO3-N and NO2-N are reduced to NH4-N by nitrate reductase and nitrite reductase (Balode., 1998; Hu., 2010). DON is not only a potential nitrogen source for primary productivity but also affects water deterioration and eutrophication (Nalepa., 1993). For the natural degradation, approximately 64% of land-based DON can degraded to DIN directly and 36% is refractory (Li., 2017).

3.2 The Total Maximum Allocated Loads of TDN on Stoichiometry

The TMALs of NH4-N, NO3-N and DON are 5700ta−1, 5800ta−1and 15600ta−1, respectively, with approximately three times of DON compared to DIN (NH4-N, NO3-N) in 2015. This indicates that the efficiency of DIN load reduction is higher than DON load reduction, while DON should be paid more attention for its increasing load proportion (Lu., 2016; Li., 2018). The TMALs are different for different outfalls in terms of spatial distribu- tion (Fig.4B), with approximately 3.5 times of DON to DIN (NH4-N, NO3-N) in LWR, 3.0 times in YR, 3.2 times in DGR, 2.4 times in MSR, 3.1 times in LSR, 2.9 times in LCR, 2.4 times in HBR and 2.0 times in TD, respectively. Based on the ratio of integration response concentrations of multiform TDN forms

Fig.3 The response concentrations of DIN (response concentration matrix) in JZB under unit loads of land-based DON, NH4-N and NO3-N. [units: μmolL−1(ta−1) −1]. 4 lines represent 4 seasons: spring, summer, autumn and winter. 3 columns represent the 3 land-based nitrogen forms input in the model.

Fig.4 The TMALs and stoichiometry (A), the TMALs of NH4-N, NO3-N and DON for outfalls (B), and monthly change in TMALs (C). The data is calculated by the 3DWQMN. Red, green and blue represent the TMAL of NH4-N, NO3-N and DON, which is abbreviated as TMALNH4-N, TMALNO3-N and TMALDONrespectively, in B and C.

There is significant relativity between the TMALs and the stoichiometry on LOIMBR of DON, NO3-N and NH4-N for all the outfalls based on their response concentration matrix (=0.54,<0.01,=24) (Fig.5A), which reflects the marine biogeochemistry capacity. The TMAL of NH4-N and NO3-N are almost equal to each other, with seasonal fluctuations of the largest values in the spring (653tm−1) and of the smallest values in the summer (241tm−1) (Fig.4C), which is mainly subject to the response concentration matrix and set water quality standards. This result indicates that the emission reduction management of land-based TDN can focus on NO3-N and NH4-N, due to their relatively more effective reduction compared to DON. However, the management of land-based DON cannot be neglected, considering the TMAL of DON on stoichiometry in the long run.

According to TMALs and the stoichiometry on LOIMBR of DON, NO3-N and NH4-N, the TMALETDNis 5340ta−1, with a decreasing sequence of TD (1120ta−1), LWR (850ta−1), LSR (820ta−1), HBR (750ta−1), LCR (600ta−1), DGR (560ta−1), MSR (380ta−1) and YR (270ta−1) in 2015 (Fig.5A). The TMALETDNof TD WWTP and LWR are more than the others due to the relatively strong hydrodynamic exchange where they are located (Liu., 2004). The TMALETDNis in the same fluctuation with TMALNin seasons (Fig.5B), decreasing in the sequence of spring> autumn > winter > summer, with average values of approximately 620, 560, 380 and 210tm−1, respectively. The significant difference in spatial-temporal distribution of TMALETDNindicates that the emission reduction management of land-based TDN should not be limited to just the forms but also more attention should be paid to time and space.

Fig.5 The TMAL of ETDN for outfalls (A) and monthly change in TMAL of ETDN (B). The data is calculated with the equation (6). Red, green and blue represent the TMAL of equivalent NH4-N, NO3-N and DON, which is abbreviated as TMALENH4-N, TMALENO3-N and TMALEDON, espectively.

Fig.6 The load of ETDN (A) and the land-based ETDN overload rate (B) in JZB in 2015. The data is calculated with the equation (8~9) based on the monitoring data in 2015. Red, green and blue represent the Load of equivalent NH4-N, NO3-N and DON, which is abbreviated as LoadENH4-N, LoadENO3-N and LoadEDON, respectively, in Fig.A. In Fig.B, red, green and blue represent the overload rate of NH4-N, NO3-N and DON, which is abbreviated as ΔF%(NH4-N), ΔF%(NO3-N) and ΔF%(DON), respectively.

3.3 Different Emission Reduction Indicators for Land-Based TDN in JZB

The LoadETDNon stoichiometry is calculated according to Eq. 11 (Fig.6A). We can see that the LoadETDNof DGR accounts for the largest proportion among the 8 outfalls, approximately 2751ta−1(accounts for 28%), while HBR, LCR, MSR, LSR, LWR, YR and TD account for 22%, 21%, 15%, 6%, 5%, 2% and 1%, respectively, in 2015. Based on the land-based pollutants overload rate (Δ%(i)) (Fig.6B), there were 4 outfalls overloaded in JZB in 2015, which lie in the head of the bay. The Δ%(i)average is 116% with a space fluctuation that decreases from DGR (391%), MSR (280%), LCR (238%) to HBR (192%). For the other 4 outfalls, YR and LSR are flat-load outfalls, whose Δ%(i)is −20% and −24%, respectively. In addition, LWR and TD are surplus-load outfalls with Δ%(i)of −36% and −91%, respectively, which might be the joint action of their low loads and strong hydrodynamic exchange at the bay mouth (Liu., 2004).

According to the identification by the overload rate, DON and NO3-N are critical forms of nitrogen to emission reduction. This can be implemented by improving the level of pollutant discharge permits for source, pollution removal by municipal sewage treatment systems and integrated management of the watershed environment. Some methods such as clean production, building or expanding municipal sewage treatment systems, upgrading sewage treatment technology, afforestation and returning farmland to forests will be helpful to reduce the loads from land.

4 Discussion

4.1 The Contribution of TDN Loads

The comprehensive treatment with industrial investigation, river reconstruction and municipal sewage treatment system construction has changed various pollutant loads from point sources to nonpoint sources. This has resulted in significant changes in the structure of the terrigenous TDN from Bay Rim, especially in the increase of proportion of DON (Fig.7). The land-based TDN emissions increased year by year, of which DIN loads have increased by approximately 13 times from 1980 to 1998 and then stabilized, while DON loads have increased by approximately 34 times since 1980 (Wang., 2006; Li., 2009; Li., 2010; Sun., 2011; Zou, 2012; Zhang., 2013; Liu., 2014). Correlatively, the structure of nitrogen in JZB has changed with the change of land-based nitrogen loads (Kong., 2016).

When the 3DWQMN was run under actual pollutant loads form 1980-2015, the DIN concentration simulated in JZB made a deviation based on the 1980. Simultaneously, the contribution of NH4-N, NO3-N and DON loads to DIN concentrations were compared according to the concentration deviation (ΔDIN) (Fig.7). The contribution of TDNloads to DIN concentration was switched in 2005, with top to down contribution order of NH4-N, NO3-N and DON before 2005, and the order of NO3-N, DON to NH4-N after 2005. The contribution of TDNto DIN concentration has changed as accompanied by the structure of TDN forms. This may be the main reason for the poor water quality at the practice of EPR.

Fig.7 The interannual variation of land-based TDN loads and contribution of TDN to DIN concentration in JZB. The columns represent the load of TDN, with gray, dark gray and white represent NO3-N, NH4-N, DON, respectively. Data were from the literature (Wang et al., 2006; Li et al., 2009; Li et al., 2010; Sun et al., 2011; Zou, 2012; Zhang et al., 2013; Liu et al., 2014). The point chart indicates the contribution of TDN to DIN concentration, which is calculated by the 3DWQMN. ΔCNO3-N, ΔCNH4-N, ΔCDON and ΔCTDN represent the concentration deviation of DIN when the pollutants are NO3-N, NH4-N, DON and TDN, respectively.

4.2 Load Carrying Analysis Based on TMAL and EPR

To compare the effects of the different nitrogen pollutant distributions in the TPLCM, three scenarios of numerical experiments with the 3DWQMN model in JZB are set. In one approach, the EPR method is used (scenario 1), which reduces pollutant loadsof NH4-N by an equal percentage (3%) for per outfall annually. In the other approaches, the same ratio (3% total load) with EPR but ‘different emission reduction’, one is the space distribution difference with respect to TDN emission reduction according to the TMAL of DIN (NH4-N, NO3-N) (scenario 2), while the other is the nitrogen form difference with respect to ETDN emission reduction according to the TMAL of ETDN, besides space distribution difference (scenario 3). Firstly, the 3DWQMN is run for 1 year under actual pollutant loads to settle the DIN to a constant value. The period of numerical experiment is set to 20 years from 2015.

TheES% was 32% in 2015, according to the survey results (Li., 2018). The ‘different emission reduction’ (scenario 2 and 3) is obviously superior than the EPR method (scenario 1), embodied in the shortening of the time span needed to reduce the DIN concentration and in the shrinking of the area of water quality over the level of grade II seawater quality (Fig.8). It needs 11 years to achieve the good water quality under scenario 2, and 7 years under scenario 3 based on the TMAL of ETDN. While, under the current EPR measures (scenario 1), after 17 years of annualES% drop of 1%, the water quality cannot be improved further (Fig.8). Under the same percentage of pollutant but different outfalls reductions (scenario 2), theES% drops significantly faster (2.3% annually) than the EPR. More efficiently, under the TMALETDNj, of the annualES% drop is 4.3% in 7 years.Thus, to achieve efficient management of water quality, it is urgent to accelerate the transformation from the ‘EPR’ to ‘different emission reduction’, which is not only reflected in expansion of the single NH4-N to ETDN including DON, NO3-N, but also in the ‘different emission reduction’ on different outfalls.

Fig.8 The interannual variation of over-level-to-area ratio (ESwq%) under 3 scenarios calculated by the 3DWQMN. Taking the loads in 2015 year as the base, 3% as the percentage of annual emission reductions. ‘EPR’ for NH4-N only (scenario 1), ‘different emission reduction’ for the overload outfalls only (scenario 2) and both overload outfalls and TDNi (scenario 3) are calculated.

5 Conclusions

Acknowledgements

This study was supported by the National Natural Science Foundation of China (No. 41676062), the NSFC-Shandong Joint Fund for Marine Ecology and Environmental Sciences (No. U1606404), the Key R&D Program of Shandong (No. 2018GHY115005), and the NSFC-Shandong Joint Fund (No. U1706215).

Borja, A., 2005. The European water framework directive: A challenge for nearshore, coastal and continental shelf research., 25 (14): 1768-1783.

Bao, X. W., Yan, J., Zhao, L., and Shi, L., 1999. Application of ecom to simulate tidal currents in Jiaozhou Bay., 23 (5): 57-60 (in Chinese with English abstract).

Beck, M. B., and Straten, G. V., 1983.. Springer Science & Business Media, New York, 105pp.

Benner, R., 2002. Chemical composition and reactivity., 384: 56-90.

Bronk, D. A., 2002. Dynamics of DON., 384: 153-247.

Bronk, D. A., See, J. H., Bradley, P., and Killberg, L., 2007. DON as a source of bioavailable nitrogen for phytoplankton., 4 (3): 283-296.

Cerco, C. F., and Cole, T., 1993. Three-dimensional eutrophication model of Chesapeake Bay., 119 (6): 1006-1025.

Codispoti, L. A., Brandes, J. A., and Christensen, J. P., 2001. The oceanic fixed nitrogen and nitrous oxide budgets: Moving targets as we enter the anthropocene?, 65 (S2): 85-105.

Dai, A. Q., Li, K. Q., and Ding, D. S., 2015. Total maximum allocated load calculation of nitrogen pollutants by linking a 3D biogeochemical-hydrodynamic model with a programming model in Bohai Sea., 111: 197-210.

Deininger, R. A., 1965. Water quality management-The planning of economically optimal pollution control systems. PhD thesis. Northweatern University, Evanston, Illinois.

Devol, A. H., and Christensen, J. P., 1993. Benthic fluxes and nitrogen cycling in sediments of the continental margin of the eastern North Pacific., 51 (2): 345-372.

Dillon, P. J., and Molot, L. A., 1990. The role of ammonium and nitrate retention in the acidification of lakes and forested catchments., 1 (1): 23-43.

Drago, M., Cescon, B., and Iovenitti, L., 2001. A three-dimen- sional numerical model for eutrophication and pollutant trans- port., 145 (1): 17-34.

Dong, Z. X., Lou, A. G., and Cui, W. L., 2010. Assessment of eutrophication of Jiaozhou Bay., 34: 35-39.

Fasham, M. J. R., Ducklow, H. W., and McKelvie, S. M., 1990. A nitrogen-based model of plankton dynamics in the oceanic mixed layer., 48 (3): 591-639.

Fennel, K., Wilkin, J., and Levin, J., 2006. Nitrogen cycling in the Middle Atlantic Bight: Results from a three-dimensional model and implications for the North Atlantic nitrogen budget., 20 (3): GB3007.

Fujiwara, O., Gnanendran, S. K., and Ohgaki, S., 1986. River quality management under stochastic stream flow., 112 (2): 185-198.

Ge, M., Wang, X. L., Yan, J., and Shi, X. Y., 2003. The calculation of environmental capacities of nutrients in the Jiaozhou Bay. PhD thesis. Ocean University of China, Qingdao.

Goldman, J. C., 1983. Kinetics of inorganic nitrogen uptake by phytoplankton. In:. Accademic Press, 233-274.

Gruber, N., and Sarmiento, J. L., 1997. Global patterns of marine nitrogen fixation and denitrification., 11 (2): 235-266.

Han, H. Y., Li, K. Q., and Wang, X. L., 2011. Environmental capacity of nitrogen and phosphorus pollutions in Jiaozhou Bay , China: Modeling and assessing., 63 (5): 262-266.

Hering, D., Borja, A., and Carstensen, J., 2010. The European water framework directive at the age of 10: A critical review of the achievements with recommendations for the future., 408 (19): 4007-4019.

Hu, L. M., Deng, S. G., and Guo, Z. G., 2009. Distribution and source of particulate organic carbon in the Bohai bay and its adjacent Bohai Sea, China., 30 (1): 39- 46.

Howarth, R. W., Sharpley, A., and Walker, D., 2002. Sources of nutrient pollution to coastal waters in the United States: Implications for achieving coastal water quality goals., 25 (4): 656-676.

Hydes, D. J., Kelly-Gerreyn, B. A., Thomson, S., Proctor, R., and Prandle, D., 1997. The biogeochemistry of nitrogen in the southern North Sea: The development of a mathematical model based on the results of the NERC-North Sea Programme surveys 1988 and 1989., DOI: 10.2307/3171673.

Jiang, Y., Xu, H. L., Hu, X. Z., Zhu, M. Z., Al-Rasheid, S., and Warren, A., 2011. An approach to analyzing spatial patterns of planktonic ciliate communities for monitoring water quality in Jiaozhou Bay, northern China., 62 (2): 227-235.

Kieber, R. J., Li, A., and Seaton, P. J., 1999. Production of nitrite from the photo degradation of dissolved organic matter in natural waters., 33 (7): 993-998.

Kong, X. J., Zhang, P., Yang, N. N., and Liang, S. K., 2016. Potential bioavailability of dissolved organic nitrogen in the discharge outlets of sewage treatment plants around the Jiaozhou Bay., 37 (3): 854-861 (in Chinese with English abstract).

Koopmans, D. J., and Bronk, D. A., 2002. Photochemical production of dissolved inorganic nitrogen and primary amines from dissolved organic nitrogen in waters of two estuaries and adjacent surficial ground waters., 26 (3): 295-304.

Li, K. Q., He, J., and Li, L. J., 2018. Linking water quality with the total load control management for nitrogen in Jiaozhou Bay , China., 85: 57-66.

Li, K. Q., Ma, Y. P., and Dai, A. Q., 2017. Degradation dynamics and bioavailability of land-based dissolved organic nitrogen in the Bohai Sea: Linking experiment with modeling., 124 (2): 856-870.

Li, K. Q., Zhang, L., Li, Y., Zhang, L. J., and Wang, X. L., 2015. A three-dimensional water quality model to evaluate the environmental capacity of nitrogen and phosphorus in Jiaozhou Bay , China., 91 (1): 306-316.

Liang, S. K., Person, S., Wu, W., Ma, Y. J., Qiao, L. L., Wang, X. H., Li, J. M., and Wang, X. L., 2015. Research and integrated coastal zone management in rapaidly developing estuarine harbours a review to inform sustainment of functions in Jiaozhou Bay, China., 116: 470-477.

Linker, L. C., Batiuk, R. A., Shenk, G. W., and Cerco, C., 2013b. Development of the Chesapeake Bay watershed total maximum daily load allocation., 49 (5): 986-1006.

Linker, L. C., Dennis, R., Shenk, G. W., Batiuk, R., Grimm, J., and Wang, P., 2013a. Computing atmospheric nutrient loads to the Chesapeake Bay watershed and tidal waters., 49 (5): 1025- 1041.

Liu, S. M., Zhang, J., Chen, H. T., and Zhang, G. S., 2005. Factors influencing nutrient dynamics in the eutrophic Jiaozhou Bay, North China., 66 (1): 66-85.

Liu, Y., Liu, C., Nelson, W. C., Shi, L., Xu, F., Liu, Y., and Yan, A., 2017. Effect of water chemistry and hydrodynamics on nitrogen transformation activity and microbial community functional potential in hyporheic zone sediment columns., 51 (9): 28pp.

Lohse, L., Malschaert, J. F. P., Helder, W., and van Raaphorst, W., 1993. Nitrogen cycling in North Sea sediments: Interaction of denitrification and nitrification in offshore and coastal areas., 101 (3): 283-296.

Loucks, D. P., Revelle, C. S., and Lynn, W. R., 1967. Linear progamming models for water pollution control., 14 (4): B-166-B-181.

Lu, D. L., 2017. The quantitative linkages between the compositions of land-based TDN and DIN concentration in Jiaozhou Bay and study on methods of the differential reduction. PhD thesis. Ocean University of China, Qingdao.

Lu, D. L., Li, K. Q., Liang, S. K., Lin, G., and Wang, X., 2016. A coastal three-dimensional water quality model of nitrogen in Jiaozhou Bay linking field experiments with modelling., 114 (1): 53-63.

Lu, D. L., Yang, N. N., Liang, S. K., Li, K. Q., and Wang, X. L., 2016. Comparison of land-based sources with ambient estuarine concentrations of total dissolved nitrogen in Jiaozhou Bay (China)., 180: 82- 90.

Lung, W. S., 2001. Water quality modeling for wasteload allocations and TMDLs.

Perciasepe, R., 1992. Nutrient reevaluation load allocations. October 14, 1992, Memorandum to the Principals’ Staff Committee Members from Secretary Robert Perciasepe, Maryland Department of the Environment, Baltimore, Maryland.

Qie, L., Chen, W. M., Wang, Z. H., Shao, Q. G., Li, X., Yuan, L. X., Hu, X. L., Zhang, W. X., and Huang, Y. H., 2012. Nitrogen-doped porous carbon nanofiber webs as anodes for lithium ion batteries with a super high capacity and rate capability., 24 (15): 2047-2050.

Qingdao Environmental Protection Bureau, 2011. ‘Equal Percent Removal Project. Qingdao. China.

Qingdao municipal Ocean and Fisheries Administration, 2002- 2017. Qingdao marine environmental bulletin for the year of 2002-2016. Qingdao. China.

Shen, Z. L., 2001. Historical changes in nutrient structure and its influences on phytoplantkon composition in Jiaozhou Bay., 52 (2): 211-224.

Shen, Z. L, Liu, Q., Wu, Y. L., and Yao, Y., 2006. Nutrient structure of seawater and ecological responses in Jiaozhou Bay , China., 69 (1): 299-307.

Simsek, H., Kasi, M., Ohm, J. B., Blonigend, M., and Khane, E., 2013. Bioavailable and biodegradable dissolved organic nitrogen in activated sludge and trickling filter wastewater treatment plants., 47 (9): 3201-3210.

Sobel, M. J., 1965. Water quality improvement programming problems., 1 (4): 477-487.

Song, J. M., Xu, Y. Y., Zhang, Y., Li, X. G., and Yuan, H. M., 2006. Progress of marine biogeochemical processes in China., 30 (2): 69-77.

Srinivasan, R., and Arnold, J. G., 1994. Integration of a basin-scale water quality model with GIS., 30 (3): 453-462.

Su, Y., 2014. Calculation of land-based pollutant allocated capacity of Shandong Province-A three-dimensional water quality simulation model-based programming methodology. PhD thesis. Ocean University of China, Qingdao.

USEPA (United States Environmental Protection Agency), 2010. Technical guidance manual for developing total maximum daily loads. U.S. Environmental Protection Agency, Office of Research and Development, Washington, D C.

Wang, S. P., 2017. Quantitative analysis of environmental quality control factors in Qingdao offshore. PhD thesis. Ocean University of China, Qingdao.

Wiegner, T. N., and Seitzinger, S. P., 2001. Photochemical and microbial degradation of external dissolved organic matter inputs to rivers., 24 (1): 27-40.

Yang, N. N., 2014. Terrigenous input, distribution and bioavail- ability of dissolved organic nitrogen in Jiaozhou Bay from 2012 to 2013. PhD thesis. Ocean University of China, Qingdao.

Zarnetske, J. P., Haggerty, R., Wondzell, S. M., and Baker, M., 2011. Dynamics of nitrate production and removal as a function of residence time in the hyporheic zone., 116, DOI: 10.1029/2010 JG001356.

Zhang, G., Zhang, J., Xu, J., and Zhang, F., 2006. Distributions, sources and atmospheric fluxes of nitrous oxide in Jiaozhou Bay., 68 (3-4): 557-566.

Zhang, J., 2007. Watersheds nutrient loss and eutrophication of the marine recipients: A case study of the Jiaozhou Bay, China., 7 (6): 583-592.

Zhang, J., 2013.. Beijing, 1-20.

Zhang, P., 2017. The technology and method of disproportionate reduction of land-based total nitrogen load and assessing the water quality effect of load reduction in Qingdao. PhD thesis. Ocean University of China, Qingdao.

Zhang, P., Su, Y., and Liang, S., 2017. Assessment of long-term water quality variation affected by high-intensity land-based inputs and land reclamation in Jiaozhou Bay, China., 75: 210-219.

# The two authors contributed equally to this work.

. E-mail: likeqiang@ouc.edu.cn

April 6, 2019;

August 1, 2019;

September 21, 2019

© Ocean University of China, Science Press and Springer-Verlag GmbH Germany 2020

(Edited by Ji Dechun)