Spatio-Temporal Variation and Controlling Factors of Water Quality in Yongding River Replenished by Reclaimed Water in Beijing, North China

water

Article Spatio-Temporal Variation and Controlling Factors of Water Quality in Yongding River Replenished by Reclaimed Water in Beijing, North China

Yilei Yu 1,2,†, Muyuan Ma 1,2,†, Fandong Zheng 3,*, Licai Liu 3, Nana Zhao 1,2, Xiaoxia Li 1,2, Yongmin Yang 4 and Jia Guo 1,2,* 1 Institute of Wetland Research, Chinese Academy of Forestry, Beijing 100091, China; [email protected] (Y.Y.); [email protected] (M.M.); [email protected] (N.Z.); [email protected] (X.L.) 2 Beijing Key Laboratory of Wetland Services and Restoration, Beijing 100091, China 3 Beijing Water Science and Technology Institute, Beijing 100048, China; [email protected] 4 Research Center on Flood and Drought Disaster Reduction of the Ministry of Water Resources, China Institute of Water Resources and Hydropower Research, Beijing 100038, China; [email protected] * Correspondence: [email protected] (F.Z.); [email protected] (J.G.) † These authors contributed equally to this work.

Received: 20 March 2017; Accepted: 19 June 2017; Published: 23 June 2017

Abstract: Reclaimed water is useful for replenishing dried up rivers in North China, although changes in water quality could be an issue. Therefore, it is essential to understand the spatio-temporal variation and the controlling factors of water quality. Samples of Yongding River water were collected seasonally, and 24 water quality parameters were analyzed in 2015. All waters were alkaline, and nitrate-nitrogen was the main form of nitrogen, while phosphorus was mostly below detection level. The water quality parameters varied in time and space. Cluster analysis showed a distinct difference between winter and the other seasons and between the natural river section and the section with reclaimed water. Based on the analysis of Gibbs plots, principal component analysis, and ionic relationships, the water chemistry was controlled by dissolution of rocks in natural river section, the quality of replenished water, the effects of dilution, and the reaction of aqueous chemistry in the reclaimed water section. The positive oxidation environment in most of the river water was conducive to the formation of nitrate-nitrogen by nitriﬁcation, and not conducive to denitriﬁcation.

Keywords: water quality; reclaimed water; controlling factors; Yongding River

1. Introduction North China is an important area of political, economic, and cultural development, while it has been facing serious water shortages. Under the influence of continual droughts, rapid socio-economic development, and a large consumption of water resources, surface runoff is significantly reduced, leading to a large number of dry rivers [1,2]. Beijing, the capital of China, has been confronted with severe water scarcity for the last 30 years. The annual average rainfall of Beijing is 584.7 mm, which is mainly concentrated from June to September [3]. The city’s total water consumption was 3.82 billion m3 in 2015, while the surface water resources were 932 million m3, and the gap was filled by excess use of groundwater [2]. Although the south-to-north Water Diversion (1 billion m3/year) largely alleviated water scarcity, water demand was not fully met. Yongding River, one of the seven major river systems in the Haihe River Basin, flows through Inner Mongolia, Shanxi, and Hebei provinces, Beijing, Tianjin, and two municipalities. Therefore, three reservoirs on the upstream of Yongding River have been used as the water supply for the city since the 1980s. The river channel about 70 km on the downstream had been dry as a result of the impoundment of the three reservoirs.

Water 2017, 9, 453; doi:10.3390/w9070453 www.mdpi.com/journal/water Water 2017, 9, 453 2 of 21

At present, reclaimed water is the “second source of water” in Beijing [3,4], the utilization rate of which has reached to 61%, with the resources of about 800 million m3, used for industrial (160 million m3)[2], agricultural (200 million m3)[5], rivers and lakes landscape (400 million m3)[6], and municipal (greenfield irrigation, car washing, road dust) (40 million m3) purposes [2]. The dry Yongding River section, whose ecological water demand was about 130 million m3, had been replenished since 2010 [7]. The reclaimed water (15 × 104 m3/day) from the nearby sewage treatment plants was used as the main supplementary water source, through 20 km of circulating pipeline and three pumping stations. The recovered surface water area of Yongding River was about 2.7 million m2. Reclaimed water is derived from treated wastewater used for a certain purpose after treatment. Due to the different treatment processes and discharge standards, some chemical components in reclaimed water are too high, which may cause the potential risk of environmental pollution. The effluents discharged from wastewater treatment plants (WWTPs) could be an important source of nutrient loading and contribute to water eutrophication [8] due to excess nutrients [9,10]. Further, it caused the risk of groundwater contamination by infiltration [11], and led to an increase in salt content in the soil, which affected plant growth and caused accumulation of heavy metals in the soil [12,13]. The characteristics of water quality could be identified by multivariate analysis and ionic relationships. The reduction of a large number of water parameters, especially the key ones, could be ascertained by principal component analysis, and the influencing factors of water chemistry could be further inferred [14–18]. Concurrently, the spatio-temporal variations of water quality could also be understood according to the cluster analysis [19–21]. In addition, the dissolved ions and their relationships were effective tools to deduce dissolution of minerals [22–25]. The comprehensive use of multiple methods would be more useful to confirm the characteristics and governing factors of water chemistry. The dry Yongding River section replenished by reclaimed water would face complex environmental issues, and therefore it is important to understand the water quality character and influencing factors. This study can further be beneficial for recovering the river by the reclaimed water. The main objectives of our study are (1) to understand the physical and chemical compositions of the river water; (2) to illustrate the spatio-temporal variations of water chemistry; and (3) to identify the controlling factors of water chemistry evolution in Yongding River replenished by the reclaimed water.

2. Materials and Methods

2.1. Study Site Yongding River is the largest river in the Haihe River Basin [26]. The study site, with an obvious continental climate, is located between the eastern humid zone and the western arid area, where average annual rainfall is about 556–560 mm, with large variations, mainly concentrated from June to September (from 1954 to 2010). The average relative humidity and annual potential evapotranspiration is 57% and 1890 mm, respectively. The terrain of Yongding River in Beijing section was high in north and low in south. The river above the Sanjiadian Reservoir is 92 km long, with an average slope of 3.1%, having a mountain erosion structural landform. The middle section (from Sanjiadian Reservoir to the Lugou Bridge), with a length of about 17 km, gradually falls from the mountains into the low mountains and plains. The downstream (from Lugou Bridge to Lianggouzhuang), with a length of 61 km, gradually widens and becomes the sandy river ﬂoodplain. The upper reaches of the Yongding River in Beijing contain mountain brown soil; the middle reaches that pass through urban area are mostly brownish yellow clay of volcanic rocks and carbonate rocks, and the main parent soil material is alluvial sediment, primarily in ﬂuvo-aquic soil. Since the 1980s, the following section of the reservoir had been dry for many years, where the main river section is about 17.4 km, and the river ecosystem had been transformed into a terrestrial ecosystem from 1980 to 2010. In order to recover the river, replenishment of the reclaimed water to the Yongding River has been carried out from 2010 [27]. The main source of replenishment is reclaimed water, followed by rainwater and surface runoff in rainy season (300 × 104 m3/year, obtained from Water 2017, 9, 453 3 of 20

Since the 1980s, the following section of the reservoir had been dry for many years, where the main river section is about 17.4 km, and the river ecosystem had been transformed into a terrestrial ecosystemWater 2017, 9 ,from 453 1980 to 2010. In order to recover the river, replenishment of the reclaimed water3 of to 21 the Yongding River has been carried out from 2010 [27]. The main source of replenishment is reclaimed water, followed by rainwater and surface runoff in rainy season (300 × 104 m3/year, obtainedBeijing Water from ConservancyBeijing Water Planning Conservancy and Design Planning Institute). and Design There Institute). is a small There tributary is a small with perennialtributary withwater perennial from the surroundingwater from residentialthe surrounding area in theresidentia left bankl area of the in river, the locatedleft bank between of the theriver, monitoring located betweenstation YD05 the andmonitoring YD06 (Figure station1), YD05 which and is mainly YD06 composed (Figure 1), of thewhich emissions is mainly of cooling composed water of from the emissionsBeijing Jingxi of cooling Gas Thermal water Powerfrom Beijing Co., Ltd. Jingxi (Beijing, Gas Thermal China). ThePower sluice Co., gate Ltd. of (Beijing, Sanjiadian China). Reservoir The sluicehas been gate closed of Sanjiadian from the 1980s,Reservoir and has the drybeen river clos sectioned from is the not 1980s, replenished and the by dry the river reservoir section (Figures is not1 replenishedand2). The illustrationby the reservoir of the (Figures replenishment 1 and 2). of The the illustration reclaimed waterof the toreplenishment Yongding River of the is reclaimed shown in waterFigure to2. Firstly,Yongding the reclaimedRiver is shown water in from Figure the surrounding2. Firstly, the WWTPs reclaimed is replenished water from to the the surrounding lower reach WWTPsof the river is replenished (station YD10), to the and lower then reach to the of upstreamthe river (station with three YD10), replenishment and then to pointsthe upstream through with the threepumping replenishment stations and points water through pipes. Thethe replenishmentpumping stations is only and fromwater May pipes. to October,The replenishment and the three is 4 4 4 3 onlyoutlet from ﬂows May (red to arrow October, positions and the in Figurethree outlet2) are fl 4ows× 10 (red, 8 ×arrow10 , positions and 3 × 10 in mFigure/day, 2) respectivelyare 4 × 104, 8(Figure × 104, 2and). The 3 × pipes104 m3 of/day, the respectively three outlet ﬂows(Figure are 2). directly The pipes set atof nearlythe three the outlet bottom flows of the are river, directly which set atonly nearly could the be bottom seen in of dry the seasons. river, which only could be seen in dry seasons.

Figure 1. The location of study site and sampling map of river water. Figure 1. The location of study site and sampling map of river water. Water 2017, 9, 453 4 of 21 Water 2017, 9, 453 4 of 20

Figure 2. The illustrationillustration ofof the the replenishment replenishment of of reclaimed reclaimed water water to to Yongding Yongding River River (a) represents(a) represents the drythe dry river river channel channel before before thereplenishment; the replenishment; (b) represents (b) represents the river the replenishedriver replenished by the by reclaimed the reclaimed water, andwater, the and background the background imagines im areagines obtained are obtained from the from Google the Google earth software). earth software).

2.2. Methods

2.2.1. Water Sampling In order to investigateinvestigate the water quality and the controlling factors of river water in Yongding River, samplessamples along along the the river river were were collected collected during du springring spring (April), (April), summer summer (July), autumn (July), (October), autumn and(October), winter and (January) winter in (January) the year in 2015. the Theyear study 2015. site The and study monitoring site and stationsmonitoring are shownstations in are Figures shown1 andin Figures2. The 1 natural and 2. riverThe natural section river (relative section to the(relative river to section the river of reclaimed section of water) reclaimed on the water) upstream on the includedupstream threeincluded stations three (YD01, stations YD02, (YD01, and YD03), YD02, and and where YD03), YD03 and is awhere reservoir. YD03 Downstream, is a reservoir. the riverDownstream, section replenished the river section by the replenished reclaimed water by the (treated reclaimed wastewater) water (treated included wastewater) seven stations included (from YD04seven tostations YD10). (from The natural YD04 riverto YD10). section The could natural be the river controlling section partcould for be further the controlling understanding part thefor riverfurther replenished understanding by reclaimed the river water. replenished by reclaimed water.

2.2.2. Analytical Techniques 2.2.2. Analytical Techniques pH, water temperature (T, ◦C), dissolved oxygen (DO), electric conductivity (EC), and pH, water temperature (T, °C), dissolved oxygen (DO), electric conductivity (EC), and oxidation-reduction potential (ORP, Eh) were measured by the portable multi-parameter water quality oxidation-reduction potential (ORP, Eh) were measured by the portable multi-parameter water analyzer (American Hach HQ-40d) produced by Hach Company (Loveland, CO, USA). Water samples quality analyzer (American Hach HQ-40d) produced by Hach Company (Loveland, CO, USA). were collected, then taken back to the laboratory under 4 ◦C cold storage, and analyzed within 24 h. Water samples were− collected, then taken back to the laboratory under 4 °C cold storage, and Bicarbonate (HCO3 ) was determined by titration under addition of sulfuric acid (0.02 mol/L), and analyzed within 24 h. Bicarbonate (HCO3−) was determined by titration under addition of sulfuric endpoint titration with methyl orange as indicator. Water samples were filtered through a 0.45 µm acid (0.02 mol/L), and endpoint titration with methyl orange as indicator. Water samples were Millipore membrane filter before analysis. The main cations including potassium (K+), sodium (Na+), filtered through a 0.45 µm Millipore membrane filter before analysis. The main cations including calcium (Ca2+), magnesium (Mg2+), and other trace metals were measured by inductively coupled potassium (K+), sodium (Na+), calcium (Ca2+), magnesium (Mg2+), and other trace metals were plasma spectroscopy (ICP-OES), with a detection limit of 0.01 mg/L and 0.01 µg/L (trace elements). measured by inductively coupled plasma spectroscopy (ICP-OES), with a detection limit of 0.01 The main anions including fluoride (F−), chloride (Cl−), sulfate (SO 2−) and nitrate (NO −) were mg/L and 0.01 µg/L (trace elements). The main anions including fluoride4 (F−), chloride (Cl−3), sulfate (SO42−) and nitrate (NO3−) were measured by ion chromatograph (Thermo Fisher ICS2100) produced by DIONEX company (Sunnyvale, CA, USA), and the detection limit was 0.01 mg/L. Ammonia Water 2017, 9, 453 5 of 21 measured by ion chromatograph (Thermo Fisher ICS2100) produced by DIONEX company (Sunnyvale, CA, USA), and the detection limit was 0.01 mg/L. Ammonia nitrogen (NH3-N), nitrite nitrogen (NO2-N) and soluble reactive phosphorus (SRP) were measured with AMS’s Smartchem 200 batch analyzer produced by Alliance company (Paris, France), with detection limits of 0.01 mg/L.

2.2.3. Data Analysis The data of cations and anions require a balance error of less than 5% before being allowed for further analysis, otherwise the data would be re-measured until the equilibrium error is met. The saturation index of rock mineral was calculated by PHREEQC software. The normality of water chemistry data was checked ﬁrst using SPSS16.0 software from International Business Machines Corporation (Armonk, NY, USA), then, the data (except pH) were log-transformed and standardized before further analysis [28,29]. Analysis of Variance (ANOVA), hierarchical clustering analysis (HCA), and principal component analysis (PCA) were all done using SPSS (SPSS16.0 for Windows). The other graphs were completed using Origin 8.5 software from OriginLab Corporation (Hampton, MA, USA). The pE represents the negative logarithm of electron activity, which is calculated in Equation (1) as [30] EhF pE = (1) 2.303RT where, F is Faraday’s constant (96.42 kJV−1 g−1 equivalent), R is the gas constant (8.314 Jmol−1 deg−1), and T is the absolute temperature in Kelvin.

3. Results and Discussion

3.1. Physical and Chemical Compositions The results of physical and chemical compositions of water samples in Yongding River are given in Table1. The pH value ranged from 8.32 to 10.32, with mean value of 9.46, which showed that all waters were alkaline. The water temperature ranged from 1 to 31.2 ◦C, with an average of 15.0 ◦C, which was mainly inﬂuenced by temperature change. The value of EC was not high, with an average of 832 µS/cm (the variation: from 391 to 1278 µS/cm). The mean value of DO was higher than 10 mg/L, and ranged from 6.49 to 18.0 mg/L. The values of ORP were mostly positive, except three samples (YD10 in spring, YD02 and YD08 in summer), which indicated oxidation water environment. Nitrogen and phosphorus are the main nutrients in river water, and the order of nitrogen forms in natural river water and water replenished by reclaimed water was the same: NO3-N > NH3-N > NO2-N, which implied that the nitrate-nitrogen was the main form. The phosphorus in most samples was below the detection level, except YD06 in spring (0.18 mg/L) and summer (0.13 mg/L). The order of dominance − 2− − of major anions in both sections was the same: HCO3 > SO4 > Cl . The order of major cations in natural river water was Na+ > Mg2+ > Ca2+ > K+, and the order in river water replenished by reclaimed water was Na+ > Ca2+ > Mg2+ > K+. The order of trace metal elements in both was consistent: Sr > Ba > Li > Fe >Se. The highest value of C.V. in natural water was SiO2, and the one in the other section was NO2-N. Some parameters in river water replenished by reclaimed water were clearly higher than the ones in the natural river water, which included DO, EC, metallic element (K+, Ca2+, Sr, Ba, Li, Se), anions − 2− (Cl , SO4 ), and nitrogen forms (NH3-N, NO2-N, NO3-N). The opposite ones included Fe and SiO2, and the other water chemistries were close in both sections. According to surface water environmental quality standards (GB3838-2002) in China [31], Class I is for the source of water, the National Nature Reserve; Class II is for centralized drinking water surface water source with the ﬁrst level protected areas, rare aquatic habitat; Class III is for centralized drinking water surface water source with second level protected areas; Class IV is for the general industrial water area and entertainment water area with indirect contact; Class V is for the agricultural water area and the general landscape requirements of the waters. Based on this standard, the average pH Water 2017, 9, 453 6 of 21

exceeded the standard value; DO met the surface water Class I; NH3-N met Class II; Total phosphorus in YD06 in Spring and Summer exceeded Class V; and F− met Class IV. The standard for reclaimed wastewater reused as scenic water (CJ/T 95-2000) in China includes two categories [32]: Class I is for entertainment landscape water with non-systemic contact with skin; Class II is for ornamental landscape water that does not involve the liquid coming into contact with skin. Most parameters met Class II.

Table 1. The results of physical and chemical composition of Yongding River water samples

Natural River Water River Replenished by Reclaimed Water Parameters Min Max Mean SD C.V. (%) Min Max Mean SD C.V. (%) pH 8.58 10.32 9.55 0.62 6.52 8.32 10.15 9.38 0.59 6.30 T(◦C) 1.0 24.0 13.7 8.48 61.96 1.4 31.2 16.4 9.15 55.95 EC (µS/cm) 391 1003 755 165.85 21.97 430 1278 910 220.77 24.27 DO (mg/L) 6.49 14.68 10.43 2.42 23.18 7.15 18.00 11.30 2.54 22.48 ORP (mv) −6.00 230.70 124.59 75.25 60.40 −121.70 215.20 124.08 84.43 68.05 NH3-N (mg/L) 0.10 0.42 0.26 0.10 39.80 0.04 1.70 0.36 0.30 81.68 NO2-N (mg/L) 0.00 0.03 0.01 0.01 95.35 0.00 0.35 0.05 0.09 168.52 NO3-N (mg/L) 0.37 1.59 0.76 0.38 50.67 0.39 14.11 2.14 3.38 157.64 F− (mg/L) 0.24 3.05 1.19 1.13 94.83 0.07 3.45 1.17 1.21 103.5 Cl− (mg/L) 28.78 160.55 60.46 39.92 66.02 19.92 214.18 77.60 54.12 69.74 2− SO4 (mg/L) 32.86 188.91 80.99 47.35 58.47 61.21 688.27 187.47 141.94 75.71 − HCO3 (mg/L) 198 415 325 74.15 22.83 116 387 230 76.39 33.22 K+ (mg/L) 1.83 6.15 3.47 1.30 37.39 3.01 17.52 9.14 4.67 51.13 Na+ (mg/L) 30.37 106.50 59.00 23.26 39.42 18.63 101.00 59.28 25.41 42.87 Ca2+ (mg/L) 4.50 15.37 9.55 2.82 29.50 12.70 78.20 31.26 15.35 49.12 Mg2+ (mg/L) 11.26 36.97 25.83 9.43 36.53 9.95 51.73 24.69 9.62 38.95 Al (mg/L) 0.01 0.07 0.03 0.02 66.95 0.00 0.06 0.02 0.01 76.92 B (mg/L) 0.05 0.17 0.10 0.04 36.32 0.02 0.15 0.07 0.03 47.14 Sr (µg/L) 140.4 296.4 185.2 43.3 23.37 214.6 3083.0 714.2 611.2 85.57 Ba (µg/L) 10.2 41.4 27.7 8.5 30.57 24.9 87.3 49.4 15.5 31.31 Fe (µg/L) 0.9 32.8 10.8 9.7 90.2 1.1 32.0 10.0 7.4 74.1 Li (µg/L) 6.1 18.7 11.5 4.1 36.2 3.9 52.9 16.6 12.6 75.7 Se (µg/L) 0.0 16.2 7.8 4.1 52.1 2.4 17.6 9.8 4.0 41.0 SiO2 (mg/L) 0.01 7.47 2.42 2.44 100.83 0.01 4.77 0.64 1.01 157.82

3.2. Spatial and Temporal Variation of Water Quality

3.2.1. pH, T, EC, DO, and ORP The variations of pH, T, EC, DO, and ORP are shown in Figure3. The pH values in summer and autumn were higher than the ones in spring and winter, except for station YD06. A signiﬁcant temporal variation was found in YD06, and the highest pH in spring, while the lowest one was in winter, and the value was close in summer and autumn. A signiﬁcant seasonal variation of water temperature was ascertained, showing highest temperature in summer, lowest in winter, and moderate in spring and autumn. Water temperature increased gradually along the river, while it was lower in the natural river section than that in the reclaimed water section. The EC in the natural river section was higher in winter and lower in summer, while the median was in spring and autumn. The peak and valley value was found in YD06 in winter (1278 µS/cm) and YD01 in summer (391 µS/cm), respectively. The EC in river section of reclaimed water was high in spring and summer, and low in autumn and winter, except for the extreme values. The peak value of DO occurred in YD06 in summer, and the next to highest one was in YD07 in autumn, while the lowest was in YD02 in summer. The obvious seasonal variation of DO in natural river section was the occurrence of high values in winter and low values in summer, while the reverse trend was observed in river section of reclaimed water. The apparent seasonal variation of ORP was: autumn > winter > spring and summer. The lowest ORP value was found in YD10 in spring (−12.7 mV), and the next lower one was in YD08 in summer (−87.3 mV). In winter, the ORP of natural river section was higher than in the section of reclaimed water, while the result was opposite in other seasons. Water 2017, 9, 453 7 of 21 Water 2017, 9, 453 7 of 20

Water 201711.0, 9, 453 32 1400 7 of 20 Spring Spring Spring Summer Summer Summer Autumn 30 Autumn Autumn 18 Winter Winter Winter 11.0 32 1400 Spring 28 Spring Spring 200 Summer 10.5 Summer Summer Autumn 3026 Autumn 1200 Autumn 18 Winter Winter Winter 16 2824 200 10.5 1200 2622 10.0 16 14 2420 1000 100

2218 10.0 S/cm) pH μ ℃ 1000 14

T(20 ) 100

9.5 16 EC( 12 ORP(mV) DO(mg/L) 18 800 14 S/cm) pH μ ℃ T( )

9.5 1612 EC( 12 0 ORP(mV) DO(mg/L) 10 9.0 800 1410 12 600 0 8 10 9.0 8 106 8.5 600 8 4 Spring -100 Spring 400 8 Summer Summer Autumn 62 6 Autumn 8.5 Winter Winter Spring 8.0 40 Spring -100 400 Summer Summer 12345678910 12345678910 12345678910 12345678910 12345678910 Autumn 2 6 Autumn monitoring stations monitoring stations monitoring stations monitoring Winter stations monitoring Winter stations 8.0 0 12345678910 12345678910 12345678910 12345678910 12345678910 FiguremonitoringFigure 3. Spatial stations3. Spatial and andmonitoring temporal temporal stations variation variationmonitoring ofof pH, stations T, T, EC, EC, DO, DO,monitoring and and ORP ORP stations in inYongding Yongdingmonitoring River. River.stations

3.2.2.3.2.2. Nitrogen NitrogenFigure and andPhosphorus 3. Spatial Phosphorus and temporal variation of pH, T, EC, DO, and ORP in Yongding River. Spatial and temporal variations of nitrogen in Yongding River are presented in Figure 4. The Spatial3.2.2. Nitrogen and temporal and Phosphorus variations of nitrogen in Yongding River are presented in Figure4. The value value of NH3-N was lower in spring, and higher in summer and autumn. The peak NH3-N value of NH3-N was lower in spring, and higher in summer and autumn. The peak NH3-N value appeared appearedSpatial in and YD06 temporal in summer, variations and the of valleynitrogen value in Yongding appeared Riverin YD05 are in presented spring. NH in Figure3-N in natural4. The in YD06 in summer, and the valley value appeared in YD05 in spring. NH -N in natural river sections valueriver sectionsof NH3-N was was mostly lower lower in spring, than the and one higher in the in section summer of reclaimed and autumn. water.3 The The peak NO NH2-N 3was-N value quite was mostlyappearedlow in river lower in YD06water, than in except the summer, one for in theand the value the section valley of YD06 of value reclaimed in springappeared (0.35 water. in mg/L).YD05 The in NOThe spring. 2NO-N3 was-N NH in3 quite -Nmost in lownaturalstations in river water,riverranged except sections from for thewas0 to value mostly2 mg/L, of lower YD06except than in for springthe YD06 one (0.35 andin the YD07. mg/L). section The Theof highestreclaimed NO3 -Nand water. in the most second The stations NO highest2-N rangedwas values quite from 0 to 2 mg/L,lowappeared in exceptriver in water,YD06 for YD06 inexcept autumn and for YD07. andthe valuewinter, The of highestrespectively YD06 in and spring. theThe second(0.35peak mg/L).values highest Theof three valuesNO 3nitrogen-N appearedin most forms stations in were YD06 in autumnrangedall found and from winter, in YD060 torespectively. 2 inmg/L, different except seasons. The for peak YD06 Phosphorus values and YD07. of only three The occurred nitrogenhighest inand forms YD06 the wereinsecond spring all highest foundand summer, values in YD06 in differentappearedwhich seasons. was in not YD06 Phosphorus described in autumn in onlythe and figure. occurredwinter, respectively in YD06 in. The spring peak and values summer, of three which nitrogen was forms not describedwere all found in YD06 in different seasons. Phosphorus only occurred in YD06 in spring and summer, in the ﬁgure. 2.0 0.4 Spring Spring Spring which was not described Summer in the figure. Summer Summer Autumn Autumn Autumn Winter Winter Winter 2.0 0.4 Spring Spring Spring Summer Summer Summer Autumn Autumn Autumn Winter Winter 14 Winter

0.3

1.5 -N(mg/L) 3 -N(mg/L)

2 8

NH 0.3 -N(mg/L) 0.2 3 NO

1.5 NO

-N(mg/L) 0.5

3 6 -N(mg/L)

2 8 NH -N(mg/L) 0.2 3 NO

0.1 NO 4 0.5 6

2 0.1 4 0.0 0 2 0.0 12345678910 12345678910 12345678910 0.0 monitoring stations monitoring stations monitoring stations 0 0.0 12345678910 12345678910 12345678910 Figuremonitoring 4. Spatial stations and temporal variationmonitoring stations of Nitrogen in Yongdingmonitoring stationsRiver.

FigureFigure 4. Spatial4. Spatial and and temporal temporal variation of of Nitrogen Nitrogen in inYongding Yongding River. River. Water 2017, 9, 453 8 of 21 Water 2017, 9, 453 8 of 20

3.2.3.3.2.3. Major Major Cations Cations and and Anions Anions SpatialSpatial and temporaltemporal variations variations of of major major cations cations and and anions anions are depictedare depicted in Figure in Figure5.F − in5. springF− in − −2− springwas quite was low, quite and low, the and content the wascontent very was different very different among monitoring among monitoring stations. Thestations. Cl andThe SOCl4 andin 2− − −2− 2− SOnatural4 in rivernatural water river were water higher were in higher summer. in summer. The peak The values peak of values Cl and ofSO Cl4 andoccurred SO4 occurred in YD04 in in YD04autumn, in andautumn, the next and highest the next appeared highest in appeared YD10 in summer in YD10 and in YD06 summer in autumn, and YD06 respectively. in autumn, The − − 2− respectively.higher value The of Cl higherin YD06 value and of YD07 Cl in in YD06 four seasons and YD07 was in another four seasons apparent was feature. another The apparent SO4 in 2− − − feature.winter was The lowerSO4 thanin winter in other was seasons. lower than The in HCO other3 seasons.in autumn The was HCO higher,3 in autumn and quite was low higher, in summer, and quiteand thelow value in summer, in natural and river the section value wasin natural higher river except section in summer. was higher The signiﬁcant except in variationsummer. ofThe K+ significantwas the peak variation value in of YD06 K+ was and the YD07 peak in value four seasons.in YD06 Naand+ wasYD07 similar in four to seasons. K+ except Na the+ was peak similar value to of KYD01+ except in autumn. the peak The value Ca2+ ofin YD01 natural in riverautumn. water The was Ca obviously2+ in natural higher river than water inthe was section obviously of reclaimed higher thanwater. in The the Casection2+ gradually of reclaimed increased water. from The YD04 Ca to2+ YD07,gradually then increased decreased from slowly YD04 from to YD08 YD07, to YD10,then decreasedand the peak slowly value from occurred YD08 to in YD10, YD06 and in summer. the peak Thevalue peak occurred of Mg 2+in YD06appeared in summer. in YD04 The insummer, peak of Mgand2+ theappeared valley occurredin YD04 in in summer, YD01 in summer.and the valley Meanwhile, occurred the in content YD01 in summer. winter was Meanwhile, high and lowthe contentin summer. in winter was high and low in summer.

4 250 700 450 Spring Spring Spring Spring Summer Summer Summer Autumn Summer 400 Winter Autumn 600 Autumn Autumn 200 Winter Winter Winter 3 500 350

300 150 400 (mg/L) - 2 (mg/L) 3 2-

4 250 F(mg/L) Cl(mg/L) 300 SO 100 HCO 200 200 1 150 50 100 100 0 0 12345678910 12345678910 12345678910 12345678910

120 Spring 18 Spring Spring Spring Summer Summer Summer 50 Summer Autumn 80 16 Autumn Autumn Autumn 100 Winter Winter Winter Winter 14 40 60 12 80

10 30 60 40 K(mg/L) Na(mg/L) Ca(mg/L) 8 Mg(mg/L)

6 20 40 4 20

2 20 10

0 0 12345678910 12345678910 12345678910 12345678910 monitoring stations monitoring stations monitoring stations monitoring stations

FigureFigure 5. 5. SpatialSpatial and and temporal temporal variation variation of of major major cations cations and and anions anions in in Yongding Yongding River. River.

3.2.4.3.2.4. Trace Trace Metallic Metallic Elements Elements Spatio-temporalSpatio-temporal variation variation of of trace trace metallic metallic elements elements is isshown shown in inFigure Figure 6. 6Similar. Similar variation variation of Al of wasAl was identified identiﬁed in four in four seasons, seasons, which which was was the the valley valley stage stage from from YD04 YD04 to to YD06. YD06. The The peak peak value value appearedappeared in in YD03 YD03 in in autumn. autumn. B B content content was was low low in in winter winter and and high high in in summer, summer, where where peak peak and and valleyvalley occurred in YD01 in autumn andand YD05YD05 inin springspringand and winter, winter, respectively. respectively. Li Li content content was was low low in inwinter, winter, where where a risinga rising stage stage occurred occurred in in YD04 YD04 and and YD05 YD05 in in all all seasons. seasons. TheThe peakpeak of SiO2 appearedappeared inin YD01 YD01 in in summer, summer, and and there there are are two two other other next next highest highest peaks in YD03 and YD08. The The peak peak and and valleyvalley of AsAs occurredoccurred in in YD04 YD04 in in winter winter and and YD03 YD03 in spring, in spring, respectively. respectively. An increasing An increasing of Sr sharply of Sr sharplyoccurred occurred from YD04 from to YD06,YD04 to and YD06, the maximum and the maximum value appeared value inappeared YD05 in spring,in YD05 while in spring, other valueswhile otherwere quitevalues low. were Seasonal quite variationlow. Seasonal of Ba wasvariation not signiﬁcant, of Ba was and not spatial significant, variations and increasedspatial variations along the increasedriver except along instation the river YD03. except The in clearstation variation YD03. The was clear that variation the increasing was that and the peak increasing value occurred and peak in value occurred in reservoir (YD03), and the content in YD01 and YD02 on upstream natural river section was very low, while higher values of YD09 and YD10 occurred downstream. Water 2017, 9, 453 9 of 21 reservoir (YD03), and the content in YD01 and YD02 on upstream natural river section was very low, Water 2017, 9, 453 9 of 20 whileWater higher 2017, values9, 453 of YD09 and YD10 occurred downstream. 9 of 20

0.08 0.20 0.06 8 Spring 0.08 0.20 Spring 0.06 Spring 8 Spring Summer Summer Summer Spring Spring SpringSummer 7 Spring Autumn Autumn Autumn Summer Summer 0.05 SummerAutumn Summer Winter Winter 7 Winter Autumn Autumn AutumnWinter Autumn 0.06 0.05 6 Winter 0.15 Winter Winter Winter 0.06 0.15 6 0.04 5 0.04 5 0.04 4

0.04 0.10 0.03 (mg/L) 2 4 B(mg/L) Al(mg/L) Li(mg/L)

0.10 0.03 (mg/L) 2 SiO 3 B(mg/L) Al(mg/L) Li(mg/L) 0.02 0.02 SiO 3 2 0.02 0.02 0.05 2 0.05 0.01 1 0.00 0.01 1 0.00 0 0.00 0.00 0 12345678910 12345678910 12345678910 12345678910 0.00 0.00 12345678910 12345678910 12345678910 12345678910 100 35 Spring Spring Spring 100 35 Summer Summer Spring 24 Spring SpringSummer Spring Autumn 3000 Autumn SpringSummer 24 Summer SummerAutumn Summer 30 Winter Winter SummerAutumn Autumn 3000 AutumnWinter 80 Autumn 30 AutumnWinter Winter Winter 80 Winter 20 25 Winter 20 25 2000 60 2000 20 16 60 g/L) g/L) g/L) g/L) 20 μ μ μ 16 μ g/L) g/L) g/L) g/L) As( Sr( μ Fe( Ba( μ μ μ 15 40 As( Sr( Fe( 12 1000 Ba( 15 40 1000 12 10 10 8 20 8 0 20 5 0 5 4 0 0 12345678910 12345678910 12345678910 12345678910 4 0 0 12345678910monitoring statiaons 12345678910monitoring statiaons 12345678910monitoring statiaons 12345678910monitoring statiaons monitoring statiaons monitoring statiaons monitoring statiaons monitoring statiaons FigureFigure 6. Spatial 6. Spatial and and temporal temporalvariation variation of of trace trace metallic metallic elements elements in Yongding in Yongding River. River. Figure 6. Spatial and temporal variation of trace metallic elements in Yongding River. 3.3. Cluster3.3. Cluster Analysis Analysis and and Spatio-Temporal Spatio-Temporal Similarity Similarity 3.3. Cluster Analysis and Spatio-Temporal Similarity In order to identify the seasonal variation, the results analyzed by cluster analysis are shown in In orderIn order to identifyto identify the the seasonal seasonal variation, the the results results analyzed analyzed by cluster by cluster analysis analysis are shown are in shown Figure 7. Four seasons were classified into two groups both in the natural river section and the in FigureFigure7. 7. Four Four seasons seasons werewere classifiedclassified into into two two groups groups both both in inthe the natural natural river river section section and andthe the reclaimed water section. Group A included spring, summer, and autumn, and Group B included reclaimedreclaimed water water section. section. Group Group A A included included spring,spring, summer, summer, and and autumn, autumn, and and Group Group B included B included winter in both sections. The distinct groups showed significant seasonal variation, and winter was winterwinter in both in both sections. sections. The The distinct distinct groups groups showed significant significant seasonal seasonal variation, variation, and andwinter winter was was particularly apparent from other seasons. The water temperatures described in Figure 3 presented particularly apparent from other seasons. The water temperatures described in Figure 3 presented particularlythe normal apparent decrease from in winter, other seasons.which was The a critical water factor temperatures influencing described some chemical in Figure reactions.3 presented the normalthe decreasenormal decrease in winter, in winter, which which was awas critical a critical factor factor influencing influencing some some chemical chemical reactions.reactions.

(a) (a)

(b) (b) Figure 7. Temporal cluster analysis in the natural river section (a) and the section replenished by Figure 7. Temporal cluster analysis in the natural river section (a) and the section replenished by Figurereclaimed 7. Temporal water (b cluster). analysis in the natural river section (a) and the section replenished by reclaimed water (b). reclaimed water (b). Therefore, the parameters with significant difference, which were identified by t-test of two Therefore, the parameters with significant difference, which were identified by t-test of two independent samples using SPSS, are shown in Table 2. In the natural river section, the ones independentTherefore, thesamples parameters using SPSS, with are significant shown in difference, Table 2. In which the natural were identifiedriver section, by thet-test ones of two independent samples using SPSS, are shown in Table2. In the natural river section, the ones including Water 2017, 9, 453 10 of 20 Water 2017, 9, 453 10 of 21 including T and As were clearly higher in Group A than in Group B, and the others including EC, DO, and NO3-N were opposite. In the reclaimed water section, the parameters T, Cl−, SO42−, Na+, TMg and2+, and As wereB in clearlyGroup higherA were in remarkably Group A thanhigher in Groupthan inB, Group and the B, in others contrary including to ORP. EC, Saturated DO, and − 2− + 2+ NOdissolved3-N were oxygen opposite. in Inpure the reclaimedwater is waterusually section, 10 mg/L the parametersunder the T, conditions Cl , SO4 ,of Na 25, Mg°C and, and an B inatmospheric Group A were pressure remarkably [33,34]. higher The thansolubility in Group of DO B, in in contrary winter towould ORP. Saturatedincrease due dissolved to the oxygen lower ◦ intemperature pure water and is usually higher 10atmospheric mg/L under pressure the conditions [33,34]. ofThe 25 DOC andcontent an atmospheric in river water pressure is influenced [33,34]. Theby solubilitymany factors, of DO insuch winter as wouldhydrodynamic increase due co tonditions, the lower phytoplankton temperature and photosynthesis, higher atmospheric and pressuredegradation [33, 34of]. organic The DO matter content [35–38]. in river Most water DO is influencedcontents of by river many water factors, in our such research as hydrodynamic were more conditions,than 10 mg/L phytoplankton [33]. This could photosynthesis, be influenced andby the degradation photosynthesis of organic of phytoplankton matter [35–38 (aquatic]. Most plant DO contentsand algae). of riverMeanwhile, water in the our organic research degradation were more in than summer 10 mg/L with [33 high]. This temperature could be influenced would consume by the photosynthesis of phytoplankton (aquatic plant and algae). Meanwhile, the organic degradation in more oxygen than in other seasons. In our study, the BOD5 was not analyzed, but the close value summer with high temperature would consume more oxygen than in other seasons. In our study, (BOD5: 20 mg/L) could be obtained from the discharge standard (CJ/T 95-2000) in China. Usually, thethe BODdecomposition5 was not analyzed, of 162g butcarbohydrates the close value needs (BOD to 5consume: 20 mg/L) 192 could mg beoxygen obtained in water. from the As discharge a result, standardthe organic (CJ/T with 95-2000) 20 mg/L in would China. consume Usually, the 23.7 decomposition mg/L of DO of[35]. 162g Therefore, carbohydrates the DO needs in summer to consume was 192lower mg than oxygen in inother water. seasons As a result,in our the study. organic DO with is the 20 mg/Lmain constituent would consume of the 23.7 oxidation mg/L of capacity, DO [35]. Therefore,therefore, thehigher DO ORP in summer occurred was in lower winter. than Although in other seasonsthe reclaimed in our study.water DOhad ishigh the maincontent constituent of major ofions, the the oxidation content capacity, would be therefore, diluted higherby the rainfall ORP occurred and surface in winter. runoff. Although the reclaimed water had high content of major ions, the content would be diluted by the rainfall and surface runoff. Table 2. The parameters with significant difference of temporal clusters in Yongding River Table 2. The parameters with significant difference of temporal clusters in Yongding River T EC DO NO3-N As

NR Group A T17.88 a 684.11 EC a 9.25 DO a 0.59 NO a3 -N12.74 a As NR GroupGroup A B 17.88 1.10a b 684.11967.33a b 13.979.25 a b 1.290.59 b a 8.0012.74 b a b b b b b Group B 1.10 T 967.33ORP13.97 Cl− SO1.2942− Na8.00 Mg B a a − a 2a− a a a RR Group A T20.80 110.18 ORP 88.50Cl 217.42SO4 65.85 Na 27.75 Mg0.08 B b b b b b b b RR GroupGroup A B 20.80 3.03a 110.18165.79a 44.8988.50 a 97.62217.42 a 39.5665.85 15.53a 27.75 0.04a 0.08 a Notes: NRGroup represents B the3.03 naturalb 165.79river water,b 44.89and RRb represents97.62 b the river39.56 sectionb 15.53 replenishedb 0.04 byb Notes:reclaimed NRrepresents water; Means the natural of riverthe water,same andparameter RR represents in two the rivergroups section with replenished the different by reclaimed letter water; are Meanssignificantly of the same different parameter (p < 0.05). in two groups with the different letter are signiﬁcantly different (p < 0.05). Spatial variation and similarity were analyzed by HCA analysis (Figure 8). All the monitoring stationsSpatial were variation classified and into similarity four groups, were which analyzed included by HCA Group analysis A (YD01, (Figure YD02,8). All and the YD03), monitoring Group stationsB (YD04 wereand YD05), classiﬁed Group into C four (YD06) groups, and which Group included D (YD07, Group YD08, A YD09, (YD01, and YD02, YD10). and The YD03), difference Group of B (YD04major ions and YD05),in the four Group groups C (YD06) was compared and Group using D (YD07, ANOVA YD08, analysis. YD09, Most and YD10).of the parameters The difference except of majorLi, Sr, ionsand inB were the four higher groups than was others. compared The major using ions ANOVA in the analysis.reclaimed Most water of section the parameters were higher except than Li, Sr,theand ones B in were the natural higher thanriver others.section. The The major small ionstributary in the from reclaimed the residential water section area on were the upstream higher than of thestation ones YD06 in the may natural lead riverto the section. increase The of the small most tributary water quality from the parameter residential (Table area 3). on As the a result, upstream the ofYD06 station belonged YD06 to may the lead separate to the group. increase of the most water quality parameter (Table3). As a result, the YD06 belonged to the separate group.

Figure 8. Spatial cluster analysis of monitoring stations in Yongding River. Figure 8. Spatial cluster analysis of monitoring stations in Yongding River. Water 2017, 9, 453 11 of 21

Table 3. Difference between spatial clusters using ANOVA analysis.

+ + 2+ − 2− − Groups EC K Na Ca Cl SO4 HCO3 NH3-N NO2-N NO3-N SiO2 B Li Sr Ba Group A 755 a 3.47 a 59.00 9.55 a 60.46 a 80.99 a 325 a 0.26 a 0.01 a 0.77 a 2.42 a 0.10 a 11.45 b 185.20 a 27.68 a Group B 1002 b 5.96 a 58.58 34.06 b 69.15 a 256.88 b 244 b 0.27 a 0.01 a 0.75 a 0.40 b 0.08 ab 33.26 a 1461.55 ab 41.28 b Group C 1141 b 15.70 ab 84.26 a 52.00 ab 116.45 b 256.97 b 290 b 0.69 b 0.21 b 9.21 b 0.82 b 0.07 ab 11.40 b 601.15 b 44.20 b Group D 806 9.09 b 53.38 b 24.67 abc 67.90 a 135.39 ab 208 ab 0.33 a 0.03 a 1.07 a 0.71 b 0.06 b 9.63 b 368.77 abc 54.71 ab Note: Means of the same parameter in four groups with the different letter are signiﬁcantly different (p < 0.05). Water 2017, 9, 453 12 of 21

The water quality parameters with significant changes among four groups are shown in Table3, and the ones with no significant changes are not listed. In the reclaimed water section, the most parameters showed the trend of first increase and then decrease, which included EC, K+, Na+, Ca2+, − − Cl , HCO3 , NH3-N, NO2-N, NO3-N, and SiO2. Group C contained only the station YDO6, where the increase of most ions was caused by the import of the small tributary on the left bank (Figures1 and2), whose flow is mainly composed of the emissions of cooling water from Beijing Jingxi Gas Thermal Power Co., Ltd. (Beijing, China). The concentration of ions in Group D was reduced due to the effect of dilution by the mixing of tributary and the river water in main channel. In addition to the effects of dilution, the reduction of NH3-N may be caused by the absorption of phytoplankton or nitrification. NO3-N had two reduction pathways of reaction, the first was the absorption of phytoplankton, and the second was transformed to N2 by the denitrification. On the other hand, according to the result of Table1 and Figure3, the oxidation environment was good for nitrification and against denitrification. Therefore, the major reactions were the absorption of phytoplankton and nitrification for nitrogen. The variation of major cations including Na+,K+, and Ca2+ may be influenced by the possible ion exchange process.

3.4. Principal Components and Controlling Factors In order to infer the controlling factors, the four spatial clusters were analyzed by principal component analysis (Table4). The components would be retained if the corresponding eigenvalue >1 [39], then the critical related parameters would be kept according to the explaining proportion >0.60 [40]. In Group A (the natural river water stations), seven principal components totally explaining 93.93% were retained. PC1 had high and positive loading with EC, DO, NO2-N, NO3-N, Se, and Sr, while negative loading with T; PC2 was highly related to K, Na+, Mg2+, B, and Li; PC3 was − 2− highly and positively correlated with Cl , SO4 , and SiO2; PC4, PC5, and PC6 had high and positive − − loading with Al, Fe, F , and Ba; PC5 was highly related to ORP and HCO3 . High and positive loading occurred with major ions and anions, as well as trace metallic elements, which showed the main chemical composition and dissolution of major minerals [41]. Loading with NO2-N, NO3-N, DO, and ORP indicated that the oxidation water environment and nitriﬁcation reaction formed the nitrate [42]. Water 2017, 9, 453 13 of 21

Table 4. Results of spatial clusters (Group A, Group B, Group C, and Group D) using principal component analysis.

Group A Group B Group C Group D Parameters PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC1 PC2 PC3 PC4 PC5 PC6 PC1 PC2 PC3 PC1 PC2 PC3 PC4 PC5 PC6 pH −0.59 0.33 0.25 −0.02 0.13 0.38 0.50 0.58 0.42 −0.17 0.26 0.08 0.62 0.94 0.20 −0.29 −0.05 0.15 0.30 0.19 −0.09 0.23 T −0.89 0.12 0.20 0.12 −0.33 −0.03 −0.03 0.59 −0.45 0.15 0.17 −0.35 0.52 0.97 0.07 0.22 0.50 −0.30 0.52 −0.40 0.04 0.26 EC 0.88 0.05 −0.35 −0.04 0.09 0.05 −0.08 0.19 −0.83 −0.28 0.33 0.24 0.15 0.25 0.48 0.48 0.01 0.65 0.06 0.06 0.42 0.31 DO 0.82 −0.03 −0.05 −0.15 0.41 0.07 0.17 −0.10 −0.26 0.87 0.13 0.31 0.18 0.81 0.26 0.53 −0.08 −0.11 0.19 0.85 −0.14 −0.16 ORP 0.21 0.20 −0.45 0.43 0.68 0.02 0.11 −0.28 0.82 0.10 0.11 0.33 0.32 −0.36 −0.92 0.15 −0.14 −0.20 −0.03 0.69 0.14 −0.63 NH3-N −0.27 −0.01 −0.02 0.77 0.01 0.43 0.26 0.32 0.64 −0.17 0.01 −0.29 0.60 0.37 0.07 0.93 −0.24 0.12 −0.19 0.87 0.09 0.22 NO2-N 0.78 −0.25 0.13 0.00 −0.21 −0.34 0.01 −0.58 −0.14 −0.09 −0.34 −0.56 −0.12 −0.14 0.99 −0.08 0.05 −0.05 0.11 −0.11 0.88 −0.04 NO3-N 0.96 0.14 0.13 −0.04 0.08 −0.08 −0.16 0.27 −0.18 0.10 0.85 0.16 0.03 −0.24 −0.68 −0.69 0.16 0.28 −0.01 −0.09 0.71 −0.09 F− −0.21 0.09 0.09 −0.14 0.03 0.94 −0.14 0.11 −0.10 0.80 0.49 −0.20 0.23 0.13 0.01 0.51 0.40 −0.18 0.57 Cl− −0.28 0.02 0.90 −0.07 −0.18 −0.01 −0.06 0.13 0.26 0.01 0.95 −0.03 0.09 0.28 −0.89 −0.36 0.29 0.17 0.84 0.05 0.29 0.05 2− SO4 −0.19 0.06 0.90 −0.10 −0.27 −0.12 −0.10 0.04 0.12 0.26 0.94 0.09 −0.03 0.06 −0.45 −0.89 0.26 0.21 0.88 −0.19 0.08 0.13 − HCO3 0.11 −0.07 −0.09 −0.06 0.97 0.04 −0.11 −0.01 0.48 0.54 0.32 0.47 0.34 0.98 0.07 −0.18 0.08 −0.39 −0.10 0.09 0.12 −0.73 K+ 0.25 0.89 0.15 0.06 −0.04 −0.04 0.09 0.86 0.20 −0.01 −0.19 0.04 0.43 −0.24 0.96 0.11 0.82 0.30 0.30 0.06 0.30 −0.07 Na+ −0.10 0.92 −0.17 −0.11 0.05 0.02 −0.05 0.97 0.04 −0.22 0.10 0.08 −0.01 −0.01 1.00 −0.09 0.86 0.21 0.15 −0.12 0.21 0.32 Ca2+ 0.55 0.47 0.35 0.22 0.26 0.02 0.47 −0.26 −0.86 0.38 −0.12 0.06 −0.08 0.00 −0.25 0.97 0.16 0.84 0.15 −0.14 −0.07 0.12 Mg2+ 0.04 0.66 −0.57 −0.20 −0.19 −0.36 −0.18 0.95 −0.14 −0.24 0.14 0.06 −0.08 0.80 0.22 0.57 0.82 0.19 0.08 −0.46 −0.23 0.03 SiO2 0.14 −0.19 0.85 −0.14 0.18 0.32 −0.02 0.33 −0.06 −0.71 0.34 −0.25 0.38 −0.87 0.10 0.49 0.39 −0.54 −0.24 −0.10 −0.17 0.63 Al −0.06 −0.07 −0.26 0.90 0.02 −0.19 −0.16 0.04 0.85 0.36 0.26 0.02 0.21 0.37 −0.10 −0.93 0.03 −0.37 0.72 0.16 −0.13 −0.13 B −0.26 0.95 0.00 −0.04 0.11 0.04 −0.03 0.94 0.11 −0.22 0.19 0.14 −0.05 0.41 −0.02 0.24 0.92 −0.02 0.13 −0.22 −0.16 0.18 Li 0.18 0.82 −0.19 −0.07 −0.09 0.35 0.16 0.90 −0.36 0.08 0.08 0.06 0.21 0.62 0.77 0.16 0.72 −0.19 0.29 −0.13 0.05 −0.03 Se 0.70 0.46 −0.20 −0.01 −0.22 −0.16 −0.03 −0.79 −0.16 −0.18 −0.33 0.36 −0.28 0.97 −0.19 0.16 −0.04 0.78 −0.05 0.28 0.12 0.25 As −0.54 0.39 −0.63 −0.03 0.20 0.17 0.25 0.12 −0.13 0.17 0.06 0.93 −0.14 −0.22 −0.13 0.97 0.83 −0.10 −0.11 0.18 0.17 −0.16 Sr 0.67 0.51 0.11 0.10 −0.05 −0.38 0.27 −0.17 −0.89 0.35 −0.07 0.08 0.03 0.12 −0.56 0.82 0.64 0.56 0.32 −0.17 0.16 0.03 Ba −0.04 −0.02 −0.25 −0.12 −0.11 −0.15 0.94 0.19 −0.02 −0.85 −0.24 −0.02 0.39 0.22 0.08 −0.15 −0.02 −0.08 −0.03 −0.47 −0.76 0.22 Fe 0.02 −0.09 0.04 0.97 0.00 −0.15 −0.09 −0.55 0.75 −0.20 0.03 0.06 −0.29 0.50 0.84 −0.22 Eigenvalues 6.74 5.13 3.83 2.61 2.50 1.59 1.09 8.24 5.83 4.87 2.09 2.08 1.14 9.54 7.33 7.12 5.27 3.76 3.13 3.09 2.58 2.28 Variance 24.54 19.17 15.71 11.03 8.48 7.95 7.06 27.10 22.56 15.30 14.74 8.57 39.76 30.55 29.68 21.96 15.65 13.03 12.86 10.73 9.49 Cumulative 24.54 43.71 59.41 70.44 78.92 86.87 93.93 27.10 49.67 64.96 79.70 88.43 97.01 39.76 70.32 100.00 21.96 37.61 50.65 63.50 74.23 83.72 Note: Variance and cumulative are in %; the components with loadings (≥0.60, bold numbers) were considered important ones. Water 2017, 9, 453 14 of 21

In Group B, six principal components (explaining 97.10%) were kept. PC1 was highly related to K+, Na+, Mg2+, and Li, while negative loading with Se; PC2 had high and positive loading with ORP and Al, while negative loading with EC, Ca, and Sr; PC3 was high and positively related to DO − and F, and negatively with Ba; PC4, PC5, and PC6 had high and positive loading with NO3-N, Cl , 2− SO4 , As, pH, T, and NH3-N. In Group C, three components totally explained the data. PC1 had high and positive loading with pH, T, and DO, while negative loading with EC; PC2 was highly and − + + negatively related to ORP, NO3-N, and Cl , and positively related to NO2-N, K , Na , and Fe. PC3 had 2+ 2− positive and high loading with NO3-N, Ca , As, and Sr, negative loading with NO3-N, SO4 , and Al. In Group D, six components (explaining 83.72%) were ascertained. PC1 had positive and high loading with K+, Na+, Mg2+, B, and As; PC2 was highly and positively related to EC, Ca2+, and Al, while it 2− − was negatively related to pH; PC3 had positive and high loading with SO4 and HCO3 ; PC4, PC5, and PC6 were highly and positively related to DO, ORP, NH3-N, NO2-N, and NO3-N, while being negatively related to Ba. The monitoring stations in Group B, Group C, and Group D all belonged to the river section replenished by reclaimed water, therefore, the high and negative loading with major cations and anions could be explained by the origin of treated wastewater (reclaimed water). Some parameters may be different among the three groups as a result of large spatial variability. The relation with ORP and nitrogen forms was different among different groups. Positive relation in Group B and Group D indicated that nitriﬁcation reaction was the controlling process of nitrogen forms. The NH3-N, NO2-N, and NO3-N in Group B were higher than the ones in Group C and Group D, which indicated that the degree of nitriﬁcation was greater [42,43].

3.5. Ionic Relationship and Hydro Geochemical Process

3.5.1. Gibbs Plot and Controlling Mechanism The mechanisms governing the water chemistry in surface water composition could be identified by the Gibbs plot [44]. Figure9 presents the Gibbs plot of spatial clusters. No samples were found in evaporation crystallization and precipitation dominant zone, and all samples lied on the rock dominance zone. TDS of samples had smaller extent variation, while Na+/Na++Ca2+ had bigger one. Group A with bigger variation of Na+/Na++Ca2+ was located to the right, which showed low content of Ca2+. On the other hand, the river section recharged by reclaimed water—including Group B, Group − − − C, and Group D—had smaller variation and was located to the left. Most values of Cl /Cl +HCO3 ranged from 0 to 0.5, and samples were located to the left. Gibbs plot shows the controlling factors of surface water chemistry (river, ocean) which are not influenced by strong human activity [44]. The water in Group A was controlled by water-rock interaction, while the river water in other groups (Group B, Group C, and Group D) was governed by the water chemistry of reclaimed water. This result also could be confirmed by the higher concentration (2.42 mg/L) of SiO2 in Group A and lower value (0.40–0.82 mg/L) in other groups. Water 2017, 9, 453 14 of 20

3.5. Ionic Relationship and Hydro Geochemical Process

3.5.1. Gibbs Plot and Controlling Mechanism The mechanisms governing the water chemistry in surface water composition could be identified by the Gibbs plot [44]. Figure 9 presents the Gibbs plot of spatial clusters. No samples were found in evaporation crystallization and precipitation dominant zone, and all samples lied on the rock dominance zone. TDS of samples had smaller extent variation, while Na+/Na++Ca2+ had bigger one. Group A with bigger variation of Na+/Na++Ca2+ was located to the right, which showed low content of Ca2+. On the other hand, the river section recharged by reclaimed water—including Group B, Group C, and Group D—had smaller variation and was located to the left. Most values of Cl−/Cl−+HCO3− ranged from 0 to 0.5, and samples were located to the left. Gibbs plot shows the controlling factors of surface water chemistry (river, ocean) which are not influenced by strong human activity [44]. The water in Group A was controlled by water-rock interaction, while the river water in other groups (Group B, Group C, and Group D) was governed by the water chemistry of reclaimed water. This result also could be confirmed by the higher concentration (2.42 mg/L) of SiO2 Water 2017, 9, 453 15 of 21 in Group A and lower value (0.40–0.82 mg/L) in other groups.

Group A Group A Group B Group B Group C Group C Group D 10000 10000 Group D Evaporation crystallization Evaporation crystallization

1000 1000

Rock dominance Rock dominance

100 100

Precipitation dominance solids (mg/L) Total dissolved Precipitation dominance Total dissolved solids (mg/L) solids dissolved Total

10 10

1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 - - - Na+/(Na++Ca2+) Cl /(Cl +HCO ) 3

FigureFigure 9. 9. GibbsGibbs plot plotof spatial of spatial clusters clusters (Group (Group A, Group A, B, Group Group B, C, Groupand Group C, and D) of Group monitoring D) of stations.monitoring stations.

3.5.2.3.5.2. Bivariate Bivariate Plot Plot and and Dissolution Dissolution Reaction Reaction TheThe ratio andand bivariate bivariate relationship relationship of differentof different ions couldions becould used be to inferused theto mineralinfer the dissolution. mineral dissolution.Figure 10 shows Figure the 10 majorshows ratio the major and relationship ratio and re oflationship samples inof Yongdingsamples in River. Yongding The CaRiver.2+/Mg The2+ Ca(Figure2+/Mg 2+10 (Figurea) of most 10a) samples of most was samples less than was 1,less which than indicated 1, which theindicated effects the of dissolutioneffects of dissolution of calcite and of calcitemagnesium and magnesium minerals, or minerals, cation exchange or cation reaction exchange causing reaction calcium causing reduction; calcium only reduction; one sample only (YD05) one samplehas a value (YD05) slightly has a greater value thanslightly 2; thegreater other than samples 2; the were other greater samples than were 1 but greater less than than 2, 1 indicating but less thanthe dissolution 2, indicating of calcitethe dissolution [22]. The of Ca calcite2++Mg 2+[22].vs. The Cation Ca2+ (Figure+Mg2+ vs.10b) Cation of samples (Figure were 10b) located of samples above werethe equilibrium located above line, the indicating equilibrium that line, the dissolutionindicating that of carbonate the dissolution and calcite of carbonate is the main and source calcite [ 24is ,the25]. mainNa++K source+ vs. [24,25]. Cl− (Figure Na++K 10+ vs.c) ofCl most− (Figure waters 10c)were of most located waters below were the located equilibrium below the line, equilibrium indicating − line,that indicating sodium and that potassium sodium and ions potassium were also affectedions were by also dissolution affected ofby salt dissolution rocks [45 of]. Thesalt rocks HCO 3[45].vs. TheNa+ HCO+Ca2+3− (Figurevs. Na+ +Ca10d)2+ of(Figure most of10d) the of samples most of were the samples located abovewere located the equilibrium above the line, equilibrium indicating line, that sodium and calcium ions were much smaller than bicarbonate, conﬁrming the dissolution of carbonate. Figure 10e shows that the natural water samples were mainly located above the equilibrium line, which showed the dissolution of carbonate, while the samples of the section replenished by reclaimed water were mainly located below the equilibrium line, indicating the effects of evaporation. The Ca2++Mg2+ − 2− vs. HCO3 +SO4 (Figure 10f) of all water samples were located below the equilibrium line, indicating that the possible ion exchange between (Na+,K+) and (Ca2+, Mg2+) or the precipitation of Ca2+ and Mg2+ [46,47]. Water 2017, 9, 453 15 of 20 indicating that sodium and calcium ions were much smaller than bicarbonate, confirming the dissolution of carbonate. Figure 10e shows that the natural water samples were mainly located above the equilibrium line, which showed the dissolution of carbonate, while the samples of the section replenished by reclaimed water were mainly located below the equilibrium line, indicating the effects of evaporation. The Ca2++Mg2+ vs. HCO3−+SO42− (Figure 10f) of all water samples were located below the equilibrium line, indicating that the possible ion exchange between (Na+, K+) and (CaWater2+,2017 Mg, 2+9,) 453 or the precipitation of Ca2+ and Mg2+ [46,47]. 16 of 21

3 8 Group A Group A (b) Group B (a) Group B Group C 6 Group C 2 Group D Group D (mol/mol) 2+ 4

/Mg 1 2+

Cations(meq/L) 1:1 Ca 2 0 12345678910 01234 stations Ca2++Mg2+(meq/L) (d) Group A (c) 6 Group B 6 Group C Group D 4 1:1

(meq/L) 4 - 3 (meq/L) - 1:1 Group A Cl Group B

2 HCO 2 Group C Group D 0 02460246 Na++K+(meq/L) Na++Ca2+(meq/L) 4 (e) (f) 6 3 1:1 1:1 (meq/L) 2+

(meq/L) 4 2 - 3

Group A +Mg Group A 2+

HCO Group B 1 Group B

2 Group C Ca Group C Group D Group D 0 0 2 4 6 8 10 12 14 16 18 20 22 246810121416182022 SO 2-+Cl-(meq/L) SO 2-+HCO -(meq/L) 4 4 3

2+ 2+ 2+ 2+ − + + Figure 10. Plot of ratioratio ofof CaCa2+/Mg2+ (a(),a ),bivariate bivariate plot plot of of cation cation vs. vs. Ca Ca2++Mg+Mg2+ (b),(b Cl), Cl− vs. vs.Na Na++K++K (c), − + 2+ − 2− − 2+ 2+ 2− − c − + 2+ d − 2− − 2+e 2+ 2− − f HCO( ), HCO3 vs.3 Navs.+Ca Na +Ca(d), HCO( ),3 HCOvs. SO3 4 vs.+Cl SO (e4), and+Cl Ca( /Mg), and vs. Ca SO/Mg4 +HCOvs.3 SO(f) 4in four+HCO groups.3 ( ) in four groups. The saturation of water samples in river water is given in Table 5. It is possible to determine the potentialThe saturationdissolution of or water precipitation samples process in river waterin the isaqueous given in solution. Table5. The It is possibleSI of gypsum to determine (CaSO4, CaSOthe potential4·2H2O) and dissolution halite (NaCl) or precipitation was negative, process while in thethe aqueous SI of calcite solution. (CaCO The3) SIand of dolomite gypsum (CaMg(CO(CaSO4, CaSO3)2) 4is·2H positive.2O) and Positive halite (NaCl) SI value was indicated negative, th whileat the the mineral SI of calcite relative (CaCO to the3) andaqueous dolomite in a supersaturated(CaMg(CO3)2) isstate; positive. a negative Positive SIvalue value indicated indicated unsaturated that the mineral state; relative a zero to thevalue aqueous indicated in a equilibriumsupersaturated state. state; This a negative indicated value the indicated dissoluti unsaturatedon process state; of gypsum a zero value and indicated salt rock, equilibrium and the precipitationstate. This indicated of calcite the and dissolution dolomite processmay occur of gypsum in both sections. and salt rock,The more and theNa precipitation+ and K+ in Figure of calcite 10c, 2− + + 2− and more dolomite SO4 may in Figure occur in10f both confirmed sections. the The potential more Na dissolutionand K inprocess Figure of 10 gypsumc, and more and salt SO4 rock.in Figure 1010a,b,df conﬁrmed showed the the potential combination dissolution of carbonat processe and of gypsum calcium and and salt magnesium rock. Figure ions 10 a,b,din the showed water, combinedthe combination with saturated of carbonate index and of calcium CaCO3 and and magnesium CaMg(CO3 ions)2, the in theprecipitation water, combined process with of carbonate saturated 2+ 2+ couldindex be of CaCOfurther3 andconfirmed, CaMg(CO which3)2, led the to precipitation less Ca and process Mg in of Figure carbonate 10f. could be further conﬁrmed, which led to less Ca2+ and Mg2+ in Figure 10f.

Table 5. Saturation index (SI) of samples in monitoring stations in Yongding River

Rock YD01 YD02 YD03 YD04 YD05 YD06 YD07 YD08 YD09 YD10

CaSO4 −3.09 −2.49 −2.87 −2.07 −2.31 −1.79 −2.21 −2.62 −2.43 −2.49 CaCO3 0.72 0.99 1.00 1.19 0.78 0.66 0.43 1.31 1.40 0.95 CaMg(CO3)2 1.62 2.30 2.16 2.42 1.23 0.80 0.47 2.40 2.67 1.64 CaSO4·2H2O −2.84 −2.24 −2.62 −1.82 −2.05 −1.53 −1.96 −2.36 −2.18 −2.24 NaCl −7.55 −6.83 −7.28 −7.06 −7.80 −6.59 −7.31 −7.85 −7.65 −7.44 Water 2017, 9, 453 16 of 20

Table 5. Saturation index (SI) of samples in monitoring stations in Yongding River

Rock YD01 YD02 YD03 YD04 YD05 YD06 YD07 YD08 YD09 YD10 CaSO4 −3.09 −2.49 −2.87 −2.07 −2.31 −1.79 −2.21 −2.62 −2.43 −2.49 CaCO3 0.72 0.99 1.00 1.19 0.78 0.66 0.43 1.31 1.40 0.95 CaMg(CO3)2 1.62 2.30 2.16 2.42 1.23 0.80 0.47 2.40 2.67 1.64 CaSO4·2H2O −2.84 −2.24 −2.62 −1.82 −2.05 −1.53 −1.96 −2.36 −2.18 −2.24 NaCl −7.55 −6.83 −7.28 −7.06 −7.80 −6.59 −7.31 −7.85 −7.65 −7.44

3.5.3. Nitrogen Forms and Redox Environment The availability of electrons for transfer between species was characterized by ORP, which could be measured in the field in surface water. The species of nitrogen in water were strongly affected by the redox environment [48,49]. In Figure 11, the ORP were mostly positive except for three samples, with positive values ranging from 47.4 to 214.6 mV, and the corresponding DO with variation from 7.21 to 18.0 mg/L. Results showed oxic condition in most of the river water. The concentrationWater 2017, 9, 453 of Fe was low in river water, with the average of 10.2 µg/L. With the increase of 17ORP, of 21 the NO3-N of most samples changed little except five samples (one in Group D and four in Group A). The oxic environment was conducive to the stable form of NO3-N, and did harm to denitrification 3.5.3. Nitrogen Forms and Redox Environment [50]. As a result, the variation of NO3-N may be controlled by mixing/dilution and the absorption of aquaticThe plants. availability of electrons for transfer between species was characterized by ORP, which could be measuredpH was inanother the field important in surface factor water. affecting The species the of fate nitrogen of NO in3-N, water and were the stronglypH-pE (computed affected by theby Equationredox environment (1)) plot could [48,49 evaluate]. In Figure the multiple11, the ORP influences were mostly of pH positive and ORP. except The forpH-pE three plot samples, for spatial with clusterspositive was values described ranging in from Figure 47.4 12. to Most 214.6 samples mV, and were the corresponding located near the DO line with (NO variation3−/NO2− from= 1), where 7.21 to the18.0 stable mg/L. Resultsnitrogen showed form oxicwas conditionnitrate (NO in most3−, +5). of theThis river result water. strongly The concentration correlated ofto Fe the was oxic low environmentin river water, in with the thenatural average river of section 10.2 µ g/L.and Withthe section the increase recharged of ORP, by reclaimed the NO3-N water, of most where samples the concentrationchanged little exceptof DO fivewas samplesmore than (one 10 in mg/L. Group Se Dveral and foursamples in Group were A).found The in oxic the environment bottom part was of diagram,conducive where to the the stable stable form nitrogen of NO3 -N,form and was did gaseous harm to nitrogen denitrification (N2, 0), [ 50which]. As meant a result, denitrification the variation didof NO not3 -Noccur. may be controlled by mixing/dilution and the absorption of aquatic plants.

300 300 300

200 200 200

100 100 100 ORP(mV) ORP(mV) ORP(mV)

0 0 0

Group A Group A Group A Group B Group B Group B Group C Group C Group C Group D Group D Group D -100 -100 -100

6 8 10 12 14 16 18 0 2 4 6 8 10 12 14 16 0 4 8 12 16 20 24 28 32 36 DO(mg/L) NO -N(mg/L) Fe(μg/L) 3

3 FigureFigure 11. 11. RelationshipRelationship of of ORP ORP and and major major redox redox species species (DO, (DO, NO NO3-N-N and and Fe) Fe) in in Yongding Yongding River. River.

pH was another important factor affecting the fate of NO3-N, and the pH-pE (computed by Equation (1)) plot could evaluate the multiple inﬂuences of pH and ORP. The pH-pE plot for spatial − − clusters was described in Figure 12. Most samples were located near the line (NO3 /NO2 = 1), where − the stable nitrogen form was nitrate (NO3 , +5). This result strongly correlated to the oxic environment in the natural river section and the section recharged by reclaimed water, where the concentration of DO was more than 10 mg/L. Several samples were found in the bottom part of diagram, where the stable nitrogen form was gaseous nitrogen (N2, 0), which meant denitriﬁcation did not occur. Water 20172017,, 99,, 453453 1817 of 2120

20 - - NO3 /NO2 =1

4 15 - - 3 O2 NO3 /NO2 =10

NO -/NO -=1 10 3 2 - NO3 H O 2 2 5 pE pE 0 Group A NH + 0 4 Group B -5 Group C Group A Group D Group B NH3 Group C -10 Group D -2

-15 024681012148910 pH pH

Figure 12. pH-pEpH-pE diagram diagram of spatial clusters of samples in Yongding River.

4. Conclusions This research focused on understanding the characteristics and major influencinginfluencing factors of river water, and and the the major major conclusions conclusions are are as as follows: follows: all allthe the waters waters are arealkaline, alkaline, and and the NO the3 NO-N was3-N wasthe main the main nitrogen nitrogen form, form, and and the the phosphorus phosphorus of ofmost most samples samples is is below below detection level.level. The spatio-temporal variation was the occurrence of peak and and valley valley of of major major water water quality parameters, which was further confirmed confirmed by by cluster cluster analysis. analysis. Four Four seasons seasons were were divided divided into into two two groups groups both both in inthe the two two river river sections, sections, and and the thesame same result result of four of four seasons seasons except except winter winter belonging belonging to one to group one group was wasobtained. obtained. All the All monitoring the monitoring stations stations were wereclassified classified into four into groups. four groups. Based Basedon the on analysis the analysis of Gibbs of Gibbsplots, plots,PCA, PCA,and ionic and ionicrelationships, relationships, it is itfound is found that that the the rock rock dominance dominance was was the the main main controlling factor, whichwhich includedincluded dissolutiondissolution ofof calcite,calcite, carbonate,carbonate, salt rock,rock, silicate,silicate, and poorpoor magnesiummagnesium minerals. The obvious variation of major cations coul couldd be be explained explained by by the the cation cation exchange exchange reaction. reaction. The oxidation environment environment in in most most river river wate waterr was was conducive conducive to tothe the stable stable form form of NO of NO3-N,3 -N,and anddid didharm harm to denitrification. to denitrification. Nitrogen Nitrogen content content of ofreclaimed reclaimed water water should should be be further further reduced reduced before discharge in order to avoid the eutrophicationeutrophication inin riverriver water.water. Acknowledgments: This research was financially supported by the National Natural Science Foundation of Acknowledgments: This research was financially supported by the National Natural Science Foundation of China (NO.China 41601037). (NO. 41601037). We wish We to thankwish to the thank three the anonymous three anon reviewersymous forreviewers their invaluable for their commentsinvaluable and comments constructive and suggestionsconstructive usedsuggestions to improve used the to qualityimprove of the the quality manuscript. of the manuscript. Author Contributions:Contributions:Yilei Yilei Yu Yu had had the the original original idea idea for the for study, the andstudy, carried and outcarried the designout the with design all co-authors. with all Yileico-authors. Yu drafted Yilei theYu manuscript,drafted the manuscript, which was revisedwhich was by allrevised authors. by all Licai authors. Liu, Yongmin Licai Liu, Yang, Yongmin Nana Yang, Zhao, Nana and XiaoxiaZhao, and Li hadXiaoxia collected Li had the collected samples the in fieldsamples and analyzedin field and the analyzed water chemistry the water in chemistry the laboratory. in the Muyuan laboratory. Ma, Jia Guo, and Fandong Zheng performed the data analysis. All authors read and approved the final manuscript. Muyuan Ma, Jia Guo, and Fandong Zheng performed the data analysis. All authors read and approved the Conflictsfinal manuscript. of Interest: The authors declare no conflicts of interest.

Conflicts of Interest: The authors declare no conflicts of interest. References

1. Jun, X. Water cycle and water resources security in North China: Problems and challenges. Adv. Geogr. 2002, 21, 517–526.

Water 2017, 9, 453 19 of 21

2. Zhou, Z.; Li, M.; Liu, W.; Jiao, Z.; Sun, F. Beijing Water Resoureces Bulletin; B.W. Authority: Beijing, China, 2015. 3. Sun, Y.; Chen, Z.; Wu, G.; Wu, Q.; Zhang, F.; Niu, Z.; Hu, H.-Y. Characteristics of water quality of municipal wastewater treatment plants in China: Implications for resources utilization and management. J. Clean. Prod. 2016, 131, 1–9. [CrossRef] 4. Yang, H.; Abbaspour, K.C. Analysis of wastewater reuse potential in Beijing. Desalination 2007, 212, 238–250. [CrossRef] 5. Wang, Z.; Li, J.; Li, Y. Using reclaimed water for agricultural and landscape irrigation in China: A review. Irrig. Drain. 2017.[CrossRef] 6. Furumai, H. Rainwater and reclaimed wastewater for sustainable urban water use. Phys. Chem. Earth Parts A/B/C 2008, 33, 340–346. [CrossRef] 7. Li, H.; Gong, X.; Gu, B.; Li, J.; Zhu, X. Study on water quality control measures of reclaimed water used in Yongding River landscape water. Beijing Water Aff. 2012, 3, 41–43. (In Chinese). 8. Qin, C.; Liu, H.; Liu, L.; Smith, S.; Sedlak, D.L.; Gu, A.Z. Bioavailability and characterization of dissolved organic nitrogen and dissolved organic phosphorus in wastewater effluents. Sci. Total Environ. 2015, 511, 47–53. [CrossRef][PubMed] 9. Xiong, J.; Wang, X.C.; Zhang, Q.; Duan, R.; Wang, N. Characteristics of a landscape water with high salinity in a coastal city of China and measures for eutrophication control. Ecol. Indic. 2016, 61, 268–273. [CrossRef] 10. Paerl, H.W.; Gardner, W.S.; McCarthy, M.J.; Peierls, B.L.; Wilhelm, S.W. Algal blooms: Noteworthy nitrogen. Science 2014, 346, 175. [CrossRef][PubMed] 11. Asano, T. Artificial Recharge of Groundwater; Elsevier: Boston, MA, USA, 2016. 12. Chen, W.; Lu, S.; Jiao, W.; Wang, M.; Chang, A.C. Reclaimed water: A safe irrigation water source? Environ. Dev. 2013, 8, 74–83. [CrossRef] 13. Qiu, Y.; Wang, H.; Li, S.; Ma, J.; Peng, Y.; Li, F. Study on effects of reclaimed water irrigation on the soil and shallow groundwater salinity. J. Tarim Univ. 2016, 28, 72–79. (In Chinese). 14. Duan, W.; Takara, K.; He, B.; Luo, P.; Nover, D.; Yamashiki, Y. Spatial and temporal trends in estimates of nutrient and suspended sediment loads in the Ishikari River, Japan, 1985 to 2010. Sci. Total Environ. 2013, 461–462, 499–508. [CrossRef][PubMed] 15. Hao, R.X.; Li, S.M.; Li, J.B.; Zhang, Q.K.; Liu, F. Water quality assessment for wastewater reclamation using principal component analysis. J. Environ. Inf. 2013, 21, 45–54. [CrossRef] 16. Subba, R.N.; Prakasa, R.J.; John, D.D.; Srinivasa, R.K.; Krishna, C. Multivariate analysis for identifying the governing factors of groundwater quality. J. Environ. Hydrol. 2001, 9, 1–9. 17. Haag, I.; Westrich, B. Processes governing river water quality identified by principal component analysis. Hydrol. Process. 2002, 16, 3113–3130. [CrossRef] 18. Petersen, W.; Bertino, L.; Callies, U.; Zorita, E. Process identification by principal component analysis of river water-quality data. Ecol. Model. 2001, 138, 193–213. [CrossRef] 19. Duan, W.; He, B.; Nover, D.; Yang, G.; Chen, W.; Meng, H.; Zou, S.; Liu, C. Water Quality Assessment and Pollution Source Identification of the Eastern Poyang Lake Basin Using Multivariate Statistical Methods. Sustainability 2016, 8, 133. [CrossRef] 20. Costa, M.; Gonçalves, A.M. Clustering and forecasting of dissolved oxygen concentration on a river basin. Stoch. Environ. Res. Risk Assess. 2010, 25, 151–163. [CrossRef] 21. Oketola, A.A.; Adekolurejo, S.M.; Osibanjo, O. Water quality assessment of River Ogun using multivariate statistical techniques. J. Environ. Prot. 2013, 4, 466–479. [CrossRef] 22. Smolders, A.J.P.; Hudson-Edwards, K.A.; Velde, G.V.D.; Roelofs, J.G.M. Controls on water chemistry of the Pilcomayo river (Bolivia, South-America). Appl. Geochem. 2004, 19, 1745–1758. [CrossRef] 23. Stallard, R.F.; Edmond, J.M. Geochemistry of the Amazon: 2. The influence of geology and weathering environment on the dissolved load. J. Geophys. Res. 1983, 88, 9671–9688. [CrossRef] 24. Liu, B.; Liu, C.Q.; Zhang, G.; Zhao, Z.Q.; Li, S.L.; Hu, J.; Ding, H.; Lang, Y.C.; Li, X.D. Chemical weathering under mid–to cool temperate and monsoon-controlled climate: A study on water geochemistry of the Songhuajiang River system, northeast China. Appl. Geochem. 2013, 31, 265–278. [CrossRef] 25. Zhu, B.; Yu, J.; Qin, X.; Rioual, P.; Zhang, Y.; Liu, Z.; Yan, M.; Li, H.; Ren, X.; Xiong, H. Identification of rock weathering and environmental control in arid catchments (northern Xinjiang) of Central Asia. J. Asian Earth Sci. 2013, 66, 277–294. [CrossRef] Water 2017, 9, 453 20 of 21

26. Jiang, B.; Wong, C.P.; Lu, F.; Ouyang, Z.; Wang, Y. Drivers of drying on the Yongding River in Beijing. J. Hydrol. 2014, 519, 69–79. [CrossRef] 27. Liang, S.; Fan, H.; Wang, L.; Yang, J.; Gao, J.; Feng, M. Study on suitability observation of ecological revetment model of Yongding River. Soil Water Conserv. Res. 2012, 19, 153–158. 28. Mujica, L.E.; Vehí, J.; Ruiz, M.; Verleysen, M.; Staszewski, W.; Worden, K. Multivariate statistics process control for dimensionality reduction in structural assessment. Mech. Syst. Signal Process. 2008, 22, 155–171. [CrossRef] 29. Reimann, C.; Filzmoser, P. Normal and lognormal data distribution in geochemistry death of a myth. Consequences for the statistical treatment of geochemical and environmental data. Environ. Geol. 2000, 39, 1001–1014. [CrossRef] 30. Cao, Y.; Tang, C.; Song, X.; Liu, C.; Zhang, Y. Characteristics of nitrate in major rivers and aquifers of the Sanjiang Plain, China. J. Environ. Monit. 2012, 14, 2624–2633. [CrossRef][PubMed] 31. Surface Water Environmental Quality Standard (gb3838-2002); China Environmental Science Press: Beijing, China, 2002. 32. Chen, L.; Yang, K.; Song, X.; He, Y.; Fan, J. Urban Wastewater Recycling Landscape Water Quality (GB/T 18921-2002); General Administration of Quality Supervision, Inspection and Quarantine of the People ‘s Republic of China: Beijing, China, 2002. 33. Truesdale, G.; Downing, A.; Lowden, G. The solubility of oxygen in pure water and sea-water. J. Chem. Technol. Biotechnol. 1955, 5, 53–62. [CrossRef] 34. Tromans, D. Temperature and pressure dependent solubility of oxygen in water: A thermodynamic analysis. Hydrometallurgy 1998, 48, 327–342. [CrossRef] 35. Zhou, W.; Yuan, X.; Long, A.; Huang, H.; Yue, W. Different hydrodynamic processes regulated on water quality (nutrients, dissolved oxygen, and phytoplankton biomass) in three contrasting waters of Hong Kong. Environ. Monit. Assess. 2014, 186, 1705–1718. [CrossRef][PubMed] 36. Radwan, M.; Willems, P.; El-Sadek, A.; Berlamont, J. Modelling of dissolved oxygen and biochemical oxygen demand in river water using a detailed and a simplified model. Int. J. River Basin Manag. 2003, 1, 97–103. [CrossRef] 37. Gelda, R.K.; Effler, S.W. A river water quality model for chlorophyll and dissolved oxygen that accommodates zebra mussel metabolism. Water Qual. Ecosyst. Model. 2000, 1, 271–309. [CrossRef] 38. Abowei, J.F.N. Salinity, dissolved oxygen, pH and surface water temperature conditions in Nkoro River, Niger Delta, Nigeria. Adv. J. Food Sci. Technol. 2010, 2, 67–71. 39. Kaiser, H.F. The varimax criteria for analytical rotation in factor analysis. Psychometrika 1958, 23, 187–200. [CrossRef] 40. Mazlum, N.; Ozer, A.; Mazlum, S.U. Interpretation of water quality data by principal components analysis. J. Eng. Environ. Sci. 1999, 23, 19–26. 41. Kim, J.-H.; Kim, R.-H.; Lee, J.; Cheong, T.-J.; Yum, B.-W.; Chang, H.-W. Multivariate statistical analysis to identify the major factors governing groundwater quality in the coastal area of Kimje, South Korea. Hydrol. Process. 2005, 19, 1261–1276. [CrossRef] 42. Strauss, E.A.; Richardson, W.B.; Bartsch, L.A.; Cavanaugh, J.C.; Bruesewitz, D.A.; Imker, H.; Heinz, J.A.; − Soballe, D.M. Nitrification in the Upper Mississippi River: Patterns, controls, and contribution to the NO3 budget. J. N. Am. Benthol. Soc. 2004, 61, 1102–1112. [CrossRef] 43. Zhao, L.; Delatolla, R.; Mohammadian, A. Nitrification kinetics & modified model for the Rideau River, Canada. Water Qual. Res. J. Can. 2013, 48, 192–201. 44. Gibbs, R.J. Mechanisms controlling world water chemistry. Science 1970, 170, 1088–1090. [CrossRef][PubMed] 45. Moon, S.; Huh, Y.; Qin, J.; Pho, N.V. Chemical weathering in the Hong (Red) River basin: Rates of silicate weathering and their controlling factors. Geochim. Cosmochim. Acta 2007, 71, 1411–1430. [CrossRef] 46. Fan, B.L.; Zhao, Z.Q.; Tao, F.X.; Liu, B.J.; Tao, Z.H.; Gao, S.; Zhang, L.H. Characteristics of carbonate, evaporite and silicate weathering in huanghe river basin: A comparison among the upstream, midstream and downstream. J. Asian Earth Sci. 2014, 96, 17–26. [CrossRef] 47. Li, J.; Yuan, G.L.; Deng, X.R.; Jing, X.M.; Sun, T.H.; Lang, X.X.; Wang, G.H. Major ion geochemistry of the Nansihu Lake basin rivers, North China: Chemical weathering and anthropogenic load under intensive industrialization. Environ. Earth Sci. 2016, 75, 453. [CrossRef] Water 2017, 9, 453 21 of 21

48. Yan, W.; Yang, L.; Wang, F.; Wang, J.; Ma, P. Riverine N2O concentrations, exports to estuary and emissions to atmosphere from the Changjiang River in response to increasing nitrogen loads. Glob. Biogeochem. Cycles 2012, 26.[CrossRef] 49. Jahangir, M.M.R.; Khalil, M.I.; Johnston, P.; Cardenas, L.M.; Hatch, D.J.; Butler, M.; Barrett, M.; O’flaherty, V.; Richards, K.G. Denitrification potential in subsoils: A mechanism to reduce nitrate leaching to groundwater. Agric. Ecosyst. Environ. 2012, 147, 13–23. [CrossRef] 50. Gomez-Velez, J.D.; Harvey, J.W.; Cardenas, M.B.; Kiel, B. Denitrification in the Mississippi River network controlled by flow through river bedforms. Nat. Geosci. 2015, 8, 941–945. [CrossRef]

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).