Wujin River is situated to the north of Taihu Lake. The river is 29 km long, about 2–3 m deep and 25–30 m wide. The river flows into Meiliang Bay of Taihu Lake as the main river of the City of Changzhou and the Wujin District, as well as into Zhushan Bay through a tributary named Yapu River. Furthermore, Wujin River is the main waterway that connects Beijing-Hangzhou Grand Canal with Taihu Lake. Both Meiliang Bay and Zhushan Bay are important bays in Taihu Lake as sources of drinking water and tourist destinations. However, these bays are also the most seriously affected by cyanobacteria blooms in Taihu Lake. The water quality of Wujin River is generally inferior to the Grade V water quality standards of China [1]. The main pollutants in the river are nitrogenous nutrients and organic pollutants, among which total nitrogen and ammonia nitrogen greatly exceed the standard. Petroleum substances, dissolved oxygen and total phosphorus fail to meet the Grade IV water quality standards in some sections of the river. In recent years, the community has paid great attention to water pollution issues associated with the Wujin River.

The Hongqi River was selected as the study area. It is one of the tributaries of the Wujin River located in the north of Taihu Lake. It runs 1.68 km between the Machi River in the west and the Wujin River in the east, and has an average width of 12.5 m, an average depth of 1.6 m, and an average annual flow velocity of 0.025 m/s. Primary sources of pollution include rural domestic sewage, industrial wastewater, farmland surface runoff, and solid waste pollution. The self-purification capability of the river is weak, partly because of the sewage emissions substantially exceed the purification capacity, and parts of the river are thus mildly eutrophic.

The study object included 1.5 km of the Hongqi River, with a watershed of 1.30 km² (Figure 1). The arable land in the area is about 0.47 km², most of which is used for paddy and vegetable cultivation, while forest and fruit trees occupy less of this area. The river is an important water source for drinking, irrigation, industry and entertainment for nearly 1,300 people. About twenty years ago, the water from the Hongqi River was drinkable after simple processing; however, rapid socioeconomic development in the area has led to increased emissions of untreated wastewater and pollutants from domestic, industrial and agricultural processes, which has resulted in decreased water quality in the river. Compared with surface water quality standards, the water quality of the Hongqi River is generally below the Grade V water quality standards [1] and the primary factors in excess are ammonia nitrogen, total nitrogen and total phosphorus.

In this study, five points P1–P5 along the Hongqi River were selected as monitoring sites (Figure 1). The distances from the monitoring sites to the downstream boundary are shown in Table 1.

The monitoring was conducted under low flow velocity conditions. In addition, data were not collected in rainy days so that the results could be fit to the steady flow model QUAL2K. Monitoring was conducted from September to December 2009 in the winter season and March to June 2010 in the spring. The following water quality and hydraulic parameters were measured: water temperature, flow velocity, depth, dissolved oxygen (DO), chemical oxygen demand (COD), ammonia nitrogen (NH₃-N), total nitrogen (TN), and total phosphorus (TP).

The water samples were collected, preserved, conveyed and monitored in accordance with the methods described in the Technical Specifications Requirements for Monitoring of Surface Water [29] and Waste Water and Water Quality Sampling-Technical Regulation of the Preservation and Handling of Samples [30]. The hydrodynamic and physical parameters, such as temperature, flow velocity and dissolved oxygen, were measured in the field [31]. Temperature and dissolved oxygen were monitored using portable sensors. Water velocity was measured using a current meter. Other parameters were measured in the laboratory. COD_Cr (the amount of oxygen required when use potassium dichromate oxidize the organic matter in 1 L sewage), ammonium nitrogen, nitrate nitrogen, total nitrogen and total phosphorus were determined using the Monitoring and Analytical Method on Water and Wastewater of China [32]. The monitoring work was conducted once a month and the average values of every four months are presented in Table 2 and Table 3.

The monitoring results indicate that the water quality of Hongqi River decreases a little from upstream to downstream. Compared with the Grade IV water quality standards (Table 4), high concentrations of nitrogen, phosphorus and chemical oxygen demand are the primary problems associated with water quality in the Hongqi River watershed. In addition, because the flow velocity of the river is slow and the sewage emissions exceed the purification capacity a lot, mild eutrophication and high turbidity remains [33,34].

QUAL2K is a one-dimensional river and stream water quality model that is an upgraded version of the QUAL2E model [35]. The QUAL2K framework, which was developed by the US Environmental Protection Agency, can simulate the migration and transformation of conventional pollutants. The model considers the stream as a one-dimensional channel with steady flow that is non-uniform and considers the influence of point source and non-point source pollution loads. The model also imitates changes with a user-opted time step inside of an hour within the daily cycle. In addition to being widely applied for the environmental management of relatively large rivers [12,18,19], the framework model also has several new features that make it applicable to shallow, upland and other rivers.

The QUAL2K framework includes the following new elements: QUAL2K uses unequally spaced reaches, and multiple loadings and withdrawals can be input to any reach. Denitrification is modeled as a first-order reaction that becomes pronounced under low oxygen concentrations. Sediment-water fluxes of dissolved oxygen and nutrients can be simulated internally rather than prescribed. That is, oxygen and nutrient fluxes are simulated as a function of settling particulate organic matter, reactions within the sediments, and concentrations of soluble forms in the overlying waters. The model explicitly simulates attached bottom algae. These algae have variable stoichiometry. Light extinction is calculated as a function of algae, detritus and inorganic solids. Both alkalinity and total inorganic carbon are simulated. QUAL2K allows the user to specify many of the kinetic parameters on a reach-specific basis [36].

QUAL2K can simulate the migration and transformation of a wide variety of constituents including dissolved oxygen, temperature, biochemical oxygen demand, organic nitrogen, ammonia nitrogen, nitrate nitrogen, total nitrogen, sediment oxygen demand, organic phosphorus, inorganic phosphorus, total phosphorus, phytoplankton and algae. The illustrations and uses of this model are described in detail in the QUAL2K user’s manual [36]. The model can also simulate some other factors, including pH, alkalinity and pathogenic bacteria. Overall, the bottom algae are indispensable for imitating shallow rivers.

For all but the bottom algae variables, a general mass balance for a component concentration c_i (Figure 2) in the reach i is written as [36]: (1)

In the above formula, c_i, Q_i, V_i, E_i, and W_i symbolize the component concentration of water quality, flow, volume, dispersion coefficient, and outer component load of reach i, respectively. S_i symbolizes the sinks and sources of the component due to a large number of transformation mechanisms and reactions in reach i. Q_out,i symbolizes flow abstraction from reach i.

According to hydrological and hydraulic conditions, locations of water quality monitoring sites, and distributions of pollution sources, the 1.5 km length of the Hongqi River was divided into three reaches, each with a length of 0.5 km. The latitude and longitude of each reach’s downstream are (31°35'35.3"N, 120°2'26.45"E), (31°35'35.1"N, 120°2'49.55"E) and (31°35'35.74"N, 120°3'6.28"E), respectively. There are 15 computational elements with a length of 100 m. Figure 3 shows the river reach with the locations of point sources of the Hongqi River. Table 5 shows the three reaches with different hydraulic characteristics.

Based on the method provided by National Water Environment Capacity Verification Manual [37] and combined with the data obtained from the local environmental management agency, the pollutants emissions were calculated [38]. The average annual domestic sewage emissions in the watershed are 90 L per person per day. It is estimated that the annual domestic sewage emissions in the drainage area are 41,719.5 t, of which COD_Cr is 7.51 t, NH₃-N is 2.09 t and TP is 0.25 t. There is no livestock or poultry breeding in the area, so agricultural non-point source emissions are primarily from farmland. It is estimated that agricultural non-point source emissions of COD_Cr are 0.14 t, while those of NH₃-N are 0.14 t, those of TN are 0.28 t and those of TP are 0.03 t per year in the drainage area. There were approximately twenty factories and enterprises in the drainage area, all of which but the Hongbo Paint Company are mechanical processing enterprises. Therefore, few pollutants are discharged into the river. It is estimated that the industrial wastewater emissions in the drainage area are 9,780 t and the COD_Cr is 20.5 t. The flow and concentration of pollution sources are shown in Table 6.

The input parameters involved in QUAL2K were temperature, flow velocity, COD_Cr, dissolved oxygen, ammonium nitrogen, nitrate nitrogen, organic nitrogen, inorganic phosphorus and organic phosphorus. The level of phytoplankton in the Hongqi River is negligible. According to field survey and hydraulic characteristics of the river, the bottom algae coverage and bottom SOD coverage were determined to be 70% and 100%, respectively.

The extent of parameters (Table 7) that QUAL2K demanded were determined from a large number of studies including documentation for the stream water quality model QUAL2E [21,35], the QUAL2K user manual [36,39] and the Environment Protection Agency guidance document [40]. The internal calculation method was applied to calculate the re-aeration rate [36,41].

The exponential model was selected for oxygen inhibition of BOD hydrolysis, COD oxidation, de-nitrification, nitrification, algae and phytoplankton respiration. The influence of wind was assumed to be ignored. There are six degradation parameters including COD oxidation rate (k_dc), ammonium nitrification rate (k_na), nitrate denitrification rate (k_dn), organic N hydrolysis rate (k_hn), organic P hydrolysis rate (k_hp), and inorganic P settling velocity rate (k_ip) were obtained by trial and error. The remaining parameters were set by the default values (Table 7) in the QUAL2K model [42,43].

The QUAL2K model has greater flexibility, which can follow the specific circumstances of users to set the parameter values and transform the simulation equation, satisfying user requirements for water quality simulation. In this study, the parameters of k_hc, k_dn, k_dt (Detritus Dissolution rate) were set to 0 and F_oxc (CBOD decay rate of rapid reaction at low dissolved oxygen conditions) was set to 1, so CBOD_f (CBOD of rapid reaction) represents the concentration of COD. The k_dc was then set as the COD comprehensive degradation coefficient; thus, QUAL2K can be used to simulate the changes of COD [36].

The monitoring data for the winter of 2009 were applied for calibration. The calculation time step was set to 5.6 min to ensure the model was maintained in the steady-state. The model was run with another completely different data set, which was set without altering the calibrated parameters, so that the ability of the calibrated model to forecast the component concentration under different circumstances could be examined. Thus, the model was utilized in the future simulation.

As shown in Figure 4, the water quality improved from the headwater to the downstream areas. The concentrations of dissolved oxygen in the Hongqi River meet the Grade III DO standards [1]. Throughout the river, the concentration of DO was greater than 5 mg/L, which indicates good water quality because it is better than the water quality objectives. Since the decomposition of pollutants consumes large amounts of dissolved oxygen, the DO is reduced. In addition, the decrease in DO concentrations was due in part to the discharge of organic pollutants by the Hongbo Paint Company, which adds high levels of organics, nitrogen substances and low DO wastewater to the Hongqi River.

The concentrations of COD, NH₃-N, TN and TP increased slightly at 0.9 km due to discharge of the local wastewater and point source pollution, which corresponds to the location of the Hongbo Paint Company. The temperature in the winter of 2009 increased from the headwater to the downstream portions of the river.

The calibration results of the QUAL2K model were in accordance with the monitoring values, with a few exceptions. For example, the simulated curves of dissolved oxygen (DO) and TP deviated slightly from the observed values. The calibrated parameters are shown in Table 7. The model was confirmed with water quality monitoring data from spring of 2010 using parameters that were calibrated based on monitoring data from the winter of 2009. The confirmation results (Figure 5) showed that the calibrated parameters are very dependable.

From Figure 5, the observed values of DO were higher or lower than the simulated values, and the observed values of TP at the location of 0.3, 0.6, and 1.4 were deviated slightly from the simulated results. The standard deviation (SD) of temperature, DO, COD, NH₃-N, TN and TP were 0.057, 0.129, 0.461, 0.020, 0.126, and 0.015, respectively. The relative standard deviation (RSD) of temperature, DO, COD, NH₃-N, TN and TP were 0.005, 0.021, 0.018, 0.008, 0.019, and 0.078, respectively. The relative standard deviation of TP was the largest in these factors. Some errors in this modeling are unavoidable because the fieldwork involved gathering a water sample at each monitoring point. Nevertheless, the simulation results were acceptable to realize water environmental management targets under the conditions of limited data.

Based on the comparative results of the pollution load emissions with the water environmental capacity of the studied river reach, the pollution load reduction required to meet the water quality objectives for the river were obtained [45]. The results are shown in Table 9.

The pollution load reductions required to satisfy the water quality objectives were calculated by subtracting the water environmental capacity from the pollution load emissions of the studied river reach [12]. Positive values indicate that the pollution load exceeds the environmental capacity and needs to be reduced, negative values indicate that the environmental capacity remains in surplus and can accommodate a greater pollution load. Based on the data in Table 9, the COD_Cr, NH₃-N, TN, and TP pollution loads of the headwater need to be reduced by −4.88 t, 1.07 t, 6.79 t, and −0.26 t, respectively, to achieve Grade IV water quality objectives. According to the water quality objectives of the Hongqi River, the pollution loads of COD_Cr, NH₃-N and TN in the river greatly exceeded the environmental capacity. These findings show that the nitrogen pollution in the river is very serious. The pollution loads of NH₃-N, TN and TP must be reduced by 50.96%, 44.11% and 22.92%, respectively, to satisfy the water quality objectives. Thus, water pollution control measures such as economic instruments or macrophytes purification [46,47], are required to carry out.