A Comprehensive Evaluation Model of Ammonia Pollution Trends in a Groundwater Source Area Along a River in Residential Areas

water

Article A Comprehensive Evaluation Model of Ammonia Pollution Trends in a Groundwater Source Area along a River in Residential Areas

Zhuoran Wang 1,* , Xiaoguang Zhao 1,2, Tianyu Xie 3, Na Wen 1 and Jing Yao 1

1 College of Geology and Environment, Xi’an University of Science & Technology, No. 58 Yanta Road, Xi’an 710054, China; [email protected] (X.Z.); [email protected] (N.W.); [email protected] (J.Y.) 2 Institute of Mine Environmental Engineering, Xi’an University of Science & Technology, No. 58 Yanta Road, Xi’an 710054, China 3 Guodu Education Science and Technology Industrial Development Zone, School of Art, Xi’an International Studies University, No. 6 Wenyuan South Road, Xi’an 710128, China; [email protected] * Correspondence: [email protected]; Tel.: +86-187-9140-8020

Abstract: In this study, a comprehensive evaluation model of ammonia pollution trends in a groundwater source area along a river in residential areas is proposed. It consists of coupling models and their interrelated models, including (i) MODFLOW and (ii) MT3DMS. The study area is laid in a plain along a river, where a few workshops operate and groundwater is heavily contaminated by domestic pollutants, agricultural pollutants, and cultivation pollutants. According to the hydrogeological conditions of the study area and the emissions of ammonia calculated in the First National Pollution Source Census Report in China, this study calibrates and verifies the prediction model. The difference between the observed water level and the calculated water level of the model is within the confidence Citation: Wang, Z.; Zhao, X.; Xie, T.; Wen, N.; Yao, J. A Comprehensive interval of the test. This means that the model is reliable and that it can truly reflect changes in the Evaluation Model of Ammonia groundwater flow field and can be directly used to simulate the migration of ammonia. The simulation Pollution Trends in a Groundwater results show that, after 20 years, the center of the ammonia pollution plume will gradually flow east Source Area along a River in along with the groundwater over time, mainly affecting the groundwater, which is less than 200 m Residential Areas. Water 2021, 13, from the river, and the ammonia content near wells at a maximum extent of less than 0.3 mg/L. 1924. https://doi.org/10.3390/ w13141924 Keywords: ammonia pollution; model; MODFLOW; pollution plume; groundwater

Academic Editors: Alexander Yakirevich and Andreas N. Angelakis 1. Introduction Received: 27 May 2021 Due to rapid growth of the global population, in order to meet growing food demands, Accepted: 5 July 2021 Published: 12 July 2021 the use of chemical fertilizers is also increasing, with nitrogen fertilizer consumption reach- ing almost 60 million t and occupying nearly 1/3 of the world total consumption in 2016,

Publisher’s Note: MDPI stays neutral which is 5 times and 2.5 times than that in the years 1985 and 2005, respectively [1–3]. with regard to jurisdictional claims in In some areas, nitrogen-containing pollutants are discharged into rivers. In this water published maps and institutional afﬁl- environment, the main types of nitrogen are ammonia, nitrite nitrogen, and nitrate nitro- iations. gen [4–6]. For example, after 20 years of Rhine River governance, although the ammonia concentration in the river water decreased by 76%, the nitrate concentration increased by approximately 35% because of the extensive use of chemical fertilizers [7]. Ammonia is toxic to aquatic organisms; it consumes oxygen in water and destroys the balance of the original ecosystem [8,9]. High concentrations of ammonia can affect human eyes, throat, Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. respiratory tract, etc. and can cause bronchitis, pneumonia, pulmonary edema, even coma, This article is an open access article and shock, posing non-carcinogenic risks to the human body [10]. distributed under the terms and Obviously, groundwater resources are one of the main sources of water for industrial conditions of the Creative Commons and agricultural purposes and in daily life [11–13]. Especially in some arid and semi-arid Attribution (CC BY) license (https:// areas, groundwater resources are the only source of water supply [14–17]. For example, in creativecommons.org/licenses/by/ China, domestic water consumption of 33% of the cities in arid areas are totally dependent 4.0/). on groundwater resources. Therefore, a large number of researchers have studied the

Water 2021, 13, 1924. https://doi.org/10.3390/w13141924 https://www.mdpi.com/journal/water Water 2021, 13, x FOR PEER REVIEW 2 of 17

Water 2021, 13, 1924 2 of 17

areas, groundwater resources are the only source of water supply [14–17]. For example, in China, domestic water consumption of 33% of the cities in arid areas are totally depend- migrationent on groundwater of pollutants resources. between Therefore, rivers and a large groundwater. number of researchers Petry demonstrated have studied that the the nitratemigration concentration of pollutants is mainly between controlled rivers and by groundwater. hydrological Petry conditions demonstrated [18]. Ohte that and the Martin ni- showedtrate concentration that the groundwater is mainly controlled nitrate concentration by hydrological distribution conditions controls [18]. Ohte seasonal and Martin nitrate variationshowed that in the the stream groundwater [19]. Lapworth nitrate concentrat showed thation distribution shallow groundwater controls seasonal is both nitrate a source andvariation a sink in for the dissolved stream [19]. N and Lapworth that deoxidation showed that conditions shallow groundwater of riparian areas is both are a important source inand N transformationa sink for dissolved [20]. N and that deoxidation conditions of riparian areas are important in NWhen transformation river water [20]. replenishes groundwater, the solute in river water enters groundwater orWhen coastal river pumping water wellsreplenishes through groundwater, a river infiltration the solute system in river by water nature. enters In this ground- process, pollutantswater or coastal in the riverpumping water wells are through affected a by river convection–diffusion, infiltration system by adsorption–desorption, nature. In this pro- biodegradation,cess, pollutants plantin the absorption,river water etc.,are affe andcted their by concentrationconvection–diffusion, is reduced adsorption–de- or eliminated; thus,sorption, the direct biodegradation, impact of aplant river absorption, pollutant on et groundwaterc., and their concentration quality is lessened. is reduced The or func- tionseliminated; related thus, to this the process direct impact are shown of a river in Figurepollutant1. Thereon groundwater are much quality research is lessened. on sewage purificationThe functions by arelated river filtrationto this process system. are For shown example, in Figure Jürgen 1. There Schubert are studiedmuch research the filtration on systemsewage of purification Rhine River by and a river found filtration that thesyst removalem. For example, rates of ammoniaJürgen Schubert and nitrite studied nitrogen the throughfiltration nitrification system of Rhine and denitrification River and found are that very the high removal [21]. Klaus-Dieterrates of ammonia Balke and et al.nitrite found thatnitrogen pollutants through are nitrification reduced by and natural denitrification attenuation are in very the processhigh [21]. of Klaus-Dieter river water infiltrationBalke et andal. found that clay that minerals, pollutants iron are hydroxide,reduced by humus,natural attenuation and underground in the process microorganisms of river water have infiltration and that clay minerals, iron hydroxide, humus, and underground microorgan- better decontamination abilities [22]. isms have better decontamination abilities [22].

Figure 1. Schematic diagram of the migration and transformation of pollutants in a river infiltration Figure 1. Schematic diagram of the migration and transformation of pollutants in a river infiltra- system. tion system. Alvarez-Cobelas et al. studied nitrogen (N) export rates from 946 rivers around the worldAlvarez-Cobelas as a function of et quantitative al. studied and nitrogen qualitative (N) export environmental rates from factors 946 rivers such aroundas land- the worlduse, population as a function density, of quantitative and dominant and qualitativehydrological environmental processes. They factors concluded such as that land-use, re- populationgional modeling density, approaches and dominant are more hydrological useful than processes. global large-scale They concluded analyses that[23]. regionalCou- modelingpled models approaches have thus arebeen more developed useful and than us globaled since large-scale the 1980s to analyses simulate [ 23N ].transfor- Coupled modelsmation haveat the thus field been scale developed (SOILN [24], and WAVE used since[25], theLEACHN 1980s to[26], simulate and CREAMS N transformation [27]) or atnitrate the field transfer scale at (SOILN the catchment [24], WAVE scale. [ 25Haiz], LEACHNhu Hu et al. [26 established], and CREAMS a two-dimensional [27]) or nitrate transferfinite element at the steady catchment flow scale.numerical Haizhu simula Hution et with al. established FEFLOW and a two-dimensionalsimulated the migra- finite elementtion and steady transformation flow numerical law of NO simulation3-, NH4+, DOC, with and FEFLOW DO solutes and in simulated a river–groundwater the migration 3− 4+ andsystem transformation under the condition law of NO when, phreatic NH , DOC, water andsupplies DO solutesriver water. in a river–groundwaterHashem A. Faidi systempointed under out that the the condition simulation when of phreaticriver–groundwater water supplies flow riverusually water. includes Hashem two parts: A. Faidi pointedflow simulation out thatthe and simulation solute transfer, of river–groundwater and water exchange flow or usually solute exchange includes twobetween parts: the flow simulationriver and groundwater and solute transfer, system. and water exchange or solute exchange between the river and groundwater system. Thus, taking the Feng River water source as the research object, this study simulates and analyzes the hydrological and water quality characteristics and the trends of groundwater ammonia pollutant migration in the study area using MODFLOW and MT3DMS in Water 2021, 13, x FOR PEER REVIEW 3 of 17

Water 2021, 13, 1924 3 of 17 Thus, taking the Feng River water source as the research object, this study simulates and analyzes the hydrological and water quality characteristics and the trends of ground- GMSwater (Groundwater ammonia pollutant Modeling migration System). in the Using study field area research using MODFLOW and consulting and MT3DMS the documents in andGMS of (Groundwater hydrogeological Modeling condition System) in the area,. Using a mathematical field research model and consulting of hydrology the and docu- water qualityments and is established of hydrogeological and used condition in GMS toin correctthe area, and a mathematical validate the mainmodel parameters. of hydrology This particularand water study quality is is beneficial established to groundwaterand used in GMS protection to correct and and provides validate athe basis main for param- scientific developmenteters. This particular and utilization study is [28beneficial–32]. to groundwater protection and provides a basis for scientific development and utilization [28–32]. 2. Materials and Methods 2.1.2. Materials Study Area and Methods 2.1. Study Area The Feng River, a water source located in the west suburb of Xi’an, Shaanxi Province, in China,The isFeng at 108 River,◦35 0a–109 water◦09 source0 east longitudelocated in andthe west 33◦50 suburb0–34◦20 of0 northXi’an, latitudeShaanxi Province, (Figure2). It boastsin China, a basin is at area108°35 of′–109°09 1380 km′ east2, an longitude average altitudeand 33°50 of′–34°20 about′ 380 north m, latitude and an average(Figure 2). annual It 2 precipitationboasts a basin of area 621 of mm. 1380 The km study, an average area of altitude the model of about is 41.94 380 km m,2. and The an groundwater average annual source 2 areaprecipitation mainly supplies of 621 mm. phreatic The stud water,y area which of the is the model only is kind 41.94 of km water. The considered groundwater in our simulation.source area Themainly groundwater supplies phreatic in this water, area is which pore wateris the only formed kind by of thewater quaternary considered system in andour cansimulation. be simulated The groundwater approximately in this as area pore is phreatic pore water water. formed by the quaternary system and can be simulated approximately as pore phreatic water.

Figure 2. Study area. Figure 2. Study area.

2.2. Numerical Simulation of Groundwater Flow 2.2.1. Generalization of Aquifers and Boundary Conditions The sand layer under the loess on the surface of the simulation area is generalized as the phreatic aquifer; the lower silty clay layer is a great impervious bed as its water permeability is weak. The vertical movement of the water ﬂow is not considered, and Water 2021, 13, x FOR PEER REVIEW 4 of 17

2.2. Numerical Simulation of Groundwater Flow Water 2021, 13, 1924 2.2.1. Generalization of Aquifers and Boundary Conditions 4 of 17 The sand layer under the loess on the surface of the simulation area is generalized as the phreatic aquifer; the lower silty clay layer is a great impervious bed as its water per- themeability water flowis weak. is generalized The vertical as movement the two-dimensional of the water flow plane is flow.not considered, Different lithologicaland the characterswater flow of is rockgeneralized mass inas thethe aquifertwo-dimensio lead tonal different plane flow. hydrogeological Different lithological parameters char- of theacters aquifer, of rock but mass with in basically the aquifer the lead same to different flow directions, hydrogeological the aquifer parameters is generalized of the aq- as the heterogeneousuifer, but with basically isotropic the medium. same flow Since direct theions, silty the clayaquifer layer is generalized is a great imperviousas the hetero- bed duegeneous to its isotropic weak water medium. permeability, Since the it silty is generalized clay layer is as a thegreat zero-flux impervious boundary. bed due The to majorits drinkingweak water water permeability, in the study it areais generalized is the shallow as the groundwater zero-flux boundary. with small The buried major depth,drinking small hydraulicwater in the gradient, study area and is weak the shallow permeability. groundwater The groundwater with small hereburied is mainlydepth, small replenished hy- bydraulic river gradient, water and and rain weak and permeability. is discharged The by groundwater outflow and here exploitation. is mainly replenished As a function by of river water and rain and is discharged by outflow and exploitation. As a function of time, time, the groundwater flow varies as rainfall duration changes, so it is generalized as the the groundwater flow varies as rainfall duration changes, so it is generalized as the un- unsteady flow [33]. steady flow [33]. The flow direction of the groundwater is roughly the same as that of the river, and both The flow direction of the groundwater is roughly the same as that of the river, and of them run from southwest to northeast. On the east and west sides of the simulation area, both of them run from southwest to northeast. On the east and west sides of the simulation the vertical line of the contour of the groundwater is viewed as the boundary impervious area, the vertical line of the contour of the groundwater is viewed as the boundary imper- to water, while the north and south sides of the simulation area are defined as the inflow vious to water, while the north and south sides of the simulation area are defined as the boundaryinflow boundary and the and outflow the outflow boundary, boundary, respectively respectively (Figure (Figure3). 3).

FigureFigure 3.3. GeneralizationGeneralization of the the boundary boundary conditions. conditions.

2.2.2.2.2.2. MeshingMeshing TheThe meshingmeshing distances of of the the model model grid grid in in the the X, X,Y, Y,and and Z directions Z directions are are6850 6850 m, m, 80608060 m,m, andand 4040 m,m, respectively.respectively. The meshing meshing cell cell is is 50 50 m m × ×5050 m. m. The The model model boundary boundary is is importedimported intointo thethe GroundwaterGroundwater Modeling Modeling System System (GMS), (GMS), displaying displaying dark dark gray gray and and light light Water 2021, 13, x FOR PEER REVIEW 5 of 17 gray.gray. TheThe darkdark gray represents represents active active cells, cells, and and the the light light gray gray represents represents inactive inactive cells cells (Figure(Figure4 ).4).

Figure 4. Meshing.Meshing.

2.2.3. Terms of Source and Sink After passing through the unsaturated zone, rain can directly enter the phreatic water. The formula of the simulated rain recharge is as follows:

Q =αPA (1)

where each part of the formula is as follows:

3 Q—quantity of rain recharges (m /d); α—coefficient of rain infiltration of each calculation partition; P—precipitation of each calculation partition (m/d); and 2 A—area of each calculation partition (m ). Due to the heterogeneity of the precipitation infiltration supply caused by different hydrogeological conditions in the study area, infiltration zoning should be used to gener- alize the precipitation infiltration supply conditions. According to the data of precipitation infiltration in the study area, the depth of the buried water level, and the lithology of the vadose zone, the zoning figure of precipitation infiltration in the study area can be drawn (Figure 5).

Figure 5. Partition of the coefficient of rain replenishment.

Water 2021, 13, x FOR PEER REVIEW 5 of 17

Water 2021, 13, 1924 5 of 17 Figure 4. Meshing.

2.2.3. Terms Terms of Source and Sink After passing through thethe unsaturatedunsaturated zone,zone, rain rain can can directly directly enter enter the the phreatic phreatic water. water.The The formula formula of the of simulatedthe simulated rain rain recharge recharge is as is follows: as follows: Q = ∑ α P A rain = i αi i (1) Qrain ∑i i iPiAi (1) where each part of the formula is as follows: Q —quantity of rain recharges (m33/d); Qrainrain—quantity of rain recharges (m /d); α αii——coefficientcoefficient of of rain rain infiltration infiltration of of each each calculation calculation partition; partition; Pi—i—precipitationprecipitation of of each each calculation calculation partition partition (m/d); (m/d); and and A 22 Ai—i—areaarea of of each each calculation calculation partition partition (m (m).). Due to the heterogeneity of the precipitationprecipitation infiltration infiltration supply caused by different hydrogeological conditions inin thethe studystudy area,area, infiltrationinfiltration zoning zoning should should be be used used to to general- gener- alizeize the the precipitation precipitation infiltration infiltration supply supply conditions. conditions. According According to theto the data data of precipitationof precipita- tioninfiltration infiltration in the in studythe stu area,dy area, the depththe depth of the of buriedthe buried water water level, level, and and the lithologythe lithology of the of thevadose vadose zone, zone, the zoningthe zoning figure figure of precipitation of precipitation infiltration infiltration in the in study the areastudy can area be can drawn be drawn(Figure (Figure5). 5).

Figure 5. PartitionPartition of of the the coefficient coefﬁcient of rain replenishment.

Combined with the buried depth of the groundwater level and the lithology of the

unsaturated zone, the precipitation infiltration coefficient in the simulation study area was finally determined as shown in Table1.

Table 1. Values of the coefﬁcient of rain replenishment in the simulation area.

Calculation Partition Parameter Value Calculation Partition Parameter Value I 0.10 II 0.08

The data show that the level of groundwater is lower than that of the river, so the groundwater is replenished by the river. The quantity of replenishment is as follows:

H2 − h2 Q = K · B (2) river 2b where each part of the formula is as follows: 3 Qriver—quantity of river recharges (m /d); K—coefficient of infiltration (m/d); B—length of the river in the study area (m); H—distance from the river level to the aquifer floor in the calculation interval (m); Water 2021, 13, 1924 6 of 17

h—distance from the phreatic water level to the aquifer ﬂoor in the calculation interval (m); and b—width of the supply zone of the river (m). The quantity of lateral runoff recharge and discharge is as follows:

= QLateral ∑i KiIiAi (3) where each part of the formula is as follows: 3 QLateral—quantity of lateral runoff recharges (m /d); Ki—coefficient of infiltration of the aquifer of part i; Ii—normal hydraulic slope of the section of part i; and 2 Ai—area of the section of the aquifer of part i (m ). The irrigation return flow recharge is as follows:

Qirrigation = Qagriβ (4)

where each part of the formula is as follows: 3 Qirrigation—quantity of well irrigation return flow (104 m /a); 3 Qagri—quantity of agricultural extraction (104 m /a); and β—coefficient of well irrigation returns flow. According to the data provided by waterworks, there are 26 total wells in the groundwater source area, 21 of which are in use, with a produced quantity of 20,000 m3/d. According to Aviriyanover, the maximum depth of evaporation from the buried phreatic Water 2021, 13, x FOR PEER REVIEW 7 of 17 water varies and ranges from 1.5 m to 4.0 m. As the depths of the buried phreatic water in the study area are all over 4 m, the evaporation capacity is viewed as zero (Figure6).

FigureFigure 6. 6. PositionsPositions of of the the river river and and wells. wells.

2.2.4.2.2.4. Hydrogeological Hydrogeological Parameter Parameter AccordingAccording to to the the data from the pumping te testst provided by waterworks, the average permeability coefficient coefﬁcient along the Feng River is 29.1429.14 m/d.m/d. The The coefficient coefﬁcient to the north of thethe Feng Feng River River with bigger peddles is 49.16 m/d and and that that in in the the south south of the Feng River with smaller pebbles is 10.32 m/d. The leakage coefficient is in the range between 3.6 × 10−7 and 2.5 × 10−6 (1/d). The simulation partitions (Figure 7) and the initial parameters of each partition are displayed in Table 2. The radius of influence to the north of the river is viewed as 237.5 m, that in the south of the river is seen as 52.3 m, and the specific yield (μ) of the aquifer is regarded as between 0.2 and 0.3.

Figure 7. Parameter partitions.

Table 2. Initial parameter values of the model.

Horizontal Permeability Coefficient Horizontal Permeability Coefficient Number Specific Yield Number Specific Yield (m/d) (m/d) 1 45.2 0.2 4 6.4 0.2 2 38.3 0.2 5 16.1 0.3 3 35.2 0.2 6 7.8 0.3

Water 2021, 13, x FOR PEER REVIEW 7 of 17

Figure 6. Positions of the river and wells.

2.2.4. Hydrogeological Parameter According to the data from the pumping test provided by waterworks, the average permeability coefficient along the Feng River is 29.14 m/d. The coefficient to the north of the Feng River with bigger peddles is 49.16 m/d and that in the south of the Feng River with smaller pebbles is 10.32 m/d. Water 2021, 13, 1924 The leakage coefficient is in the range between 3.6 × 10−7 and 2.5 × 10−6 (1/d).7 ofThe 17 simulation partitions (Figure 7) and the initial parameters of each partition are displayed in Table 2. TheThe leakageradius of coefﬁcient influence isto inthe the north range of betweenthe river 3.6is viewed× 10−7 asand 237.5 2.5 m,× 10that−6 in(1/d). the south The simulationof the river partitionsis seen as 52.3 (Figure m, and7) and the the specific initial yield parameters (μ) of the of aquifer each partition is regarded are displayedas between in0.2 Table and 20.3..

FigureFigure 7.7. ParameterParameter partitions.partitions.

TableTable 2.2.Initial Initial parameterparameter valuesvalues of of the the model. model. Horizontal Permeability Coefficient Horizontal Permeability Coefficient Number Horizontal Permeability SpecificSpecific Yield Number Horizontal Permeability SpecificSpecific Yield Number (m/d) Number (m/d) Coefficient (m/d) Yield Coefficient ((m/d) Yield 1 45.2 0.2 4 6.4 0.2 2 138.3 45.2 0.2 0.2 5 416.1 6.4 0.2 0.3 3 235.2 38.3 0.2 0.2 6 57.8 16.1 0.3 0.3 3 35.2 0.2 6 7.8 0.3

The radius of inﬂuence to the north of the river is viewed as 237.5 m, that in the south of the river is seen as 52.3 m, and the speciﬁc yield (µ) of the aquifer is regarded as between 0.2 and 0.3.

2.2.5. Mathematical Model

 ∂ ∂H ∂ ∂H ∂H  ε µ  K[H − Z(x, y)] + K[H − Z(x, y)] + = (x, y) ∈ Ω, t > 0;  ∂x ∂x ∂y ∂y ∂t  H(x, y, t)|t=0 = H0(x, y)(x, y) ∈ Ω, t > 0; (5)   ∂H  K = q(x, y)(x, y) ∈ Γ , t > 0.  n ∂n Γ2 2

where each part of the model is as follows: Ω—vadose zone; H—height of the groundwater level (m); K—hydraulic conductivity of the aquifer in horizontal direction (m/d); H0—initial flow field (m); ε—terms of source and sink of the aquifer (m/d); Γ2—second-class boundary of the vadose zone; n—normal direction of the boundary surface; ∂H ∂n —derivative of H along the outer n (dimensionless); 2 q—single-width flow on Γ2 (m /d), with its inflow positive and its outflow negative; and Z(x, y)—elevation of the aquifer floor (m). Water 2021, 13, x FOR PEER REVIEW 8 of 17

2.2.5. Mathematical Model ∂ ∂H ∂ ∂H ∂H ⎧ KH − Zx, y + KH − Zx, y +ε=μ x, y ∈Ω,t0; ⎪∂x ∂x ∂y ∂y ∂t Hx, y, t| =H0x, y x, y ∈Ω,t0; (5) ⎨ t=0 ⎪ ∂H Kn | =qx, y x, y ∈г2,t 0. ⎩ ∂n г2

where each part of the model is as follows: Ω—vadose zone; H—height of the groundwater level (m); K—hydraulic conductivity of the aquifer in horizontal direction (m/d); H0—initial flow field (m); ε—terms of source and sink of the aquifer (m/d); Г2—second-class boundary of the vadose zone; n—normal direction of the boundary surface; Water 2021, 13, 1924 —derivative of H along the outer n (dimensionless); 8 of 17 q—single-width flow on Г2 (m2/d), with its inflow positive and its outflow negative; and Zx, y—elevation of the aquifer floor (m).

2.2.6.2.2.6. SolvingSolving ProcessProcess TheThe MODFLOWMODFLOW solvingsolving processprocess is is shown shown in in Figure Figure8 .8.

FigureFigure 8.8.MODFLOW MODFLOW solvingsolving process.process.

2.2.7.2.2.7. TemporalTemporal Discretization Discretization TheThe periodperiod from from 1 1 JanuaryJanuary 20202020 toto 3131 DecemberDecember 20202020 waswas selectedselected toto verifyverify thethe model.model. EachEach correspondingcorresponding calendarcalendar monthmonth waswas takentaken asas aa stressstress period,period, andand then,then, thethe wholewhole simulationsimulation period period was was divided divided into into 12 12 stress stress periods. periods. The The step step length length of of each each stress stress period period waswas 1010 days. days. AfterAfter thethe modelmodel waswas identifiedidentified andand verified,verified, the the corresponding corresponding information information waswas inputinput andand output.output. Then, the the dynamic dynamic changes changes in in the the groundwater groundwater level level in in the the next next 20 20years years from from 1 January 1 January 2021 2021 to 31 to December 31 December 2040 2040 in the in thegroundwater groundwater source source area area were were sim- simulatedulated and and predicted. predicted. 2.2.8. Run the Model Under the condition of normal pumping in the groundwater source area, Figure9 stimulates the contours of water for different periods of 3 years, 5 years, 10 years, 13 years, 15 years, and 20 years. The overall changes in the groundwater flow field in the next 20 years can be seen. Influenced by precipitation, evaporation, the amount of pumping, and local hydrogeological conditions, the groundwater level in the study area changes obviously. Additionally, the long-term exploitation of groundwater will lead to the cone of depression of the groundwater level around the exploitation site. Water 2021, 13, x FOR PEER REVIEW 9 of 17

2.2.8. Run the Model Under the condition of normal pumping in the groundwater source area, Figure 9 stimulates the contours of water for different periods of 3 years, 5 years, 10 years, 13 years, 15 years, and 20 years. The overall changes in the groundwater flow field in the next 20 years can be seen. Influenced by precipitation, evaporation, the amount of pumping, and local hydrogeological conditions, the groundwater level in the study area changes obvi- Water 2021, 13, 1924 ously. Additionally, the long-term exploitation of groundwater will lead to the cone9 of 17 of depression of the groundwater level around the exploitation site.

(a) (b)

(e) (f)

Figure 9. Distribution of water level and ﬂow ﬁeld at different simulation times. (a) Simulates the water level after 3 years under nomal pumping conditions (b) Simulates the water level after 5 years under nomal pumping conditions (c) Simulates the water level after 10 years under nomal pumping conditions (d) Simulates the water level after 13 years under nomal pumping conditions (e) Simulates the water level after 15 years under nomal pumping conditions (f) Simulates the water level after 20 years under nomal pumping conditions. Water 2021, 13, x FOR PEER REVIEW 10 of 17

Figure 9. Distribution of water level and flow field at different simulation times.(a) Simulates the water level after 3 years under nomal pumping conditions (b) Simulates the water level after 5 years under nomal pumping conditions (c) Simu- lates the water level after 10 years under nomal pumping conditions (d) Simulates the water level after 13 years under Waternomal2021, 13 pumping, 1924 conditions (e) Simulates the water level after 15 years under nomal pumping conditions (f) Simulates the10 of 17 water level after 20 years under nomal pumping conditions.

2.2.9. Model Identification and Validation 2.2.9. Model Identification and Validation The simulation controls the fitting precision by virtue of controlling and observing The simulation controls the fitting precision by virtue of controlling and observing the the dynamic changes in the groundwater level and collects data about dynamic changes dynamic changes in the groundwater level and collects data about dynamic changes in the in the groundwater level from 2019 to 2020 to identify the model. There are six wells for groundwater level from 2019 to 2020 to identify the model. There are six wells for water waterlevel observationlevel observation that can that offer can offer data, data, and alland of all them of them are locatedare located on the on eastthe east bank bank of the of theFeng Feng River. River. According According to to the the position position of of the the error error bars bars of of the the datadata andand thethe colors, which showshow how how the the model model is is adjusted, adjusted, when when the the erro errorr bars bars are are above above the the observed observed value, value, they they indicateindicate that that the the actual actual water water level level is is lower lower than than the the calculated calculated water water level level while, while, when when theythey are are below thethe observedobserved value,value, they they show show that that the the actual actual water water level level is higheris higher than than the thecalculated calculated water water level. level. When When the datathe data for the for calculated the calculated water water level andlevel the and actual the actual water waterlevel arelevel within are within the confidence the confidence interval, interval, the error the bars error are bars filled are with filled green. with When green. the When data theare data above are the above confidence the confidence interval butinterval by no bu moret by thanno more 200%, than the 200%, errorbars the error are filled bars with are filledorange. with Additionally, orange. Additionally, when it is abovewhen 200%,it is ab theove error200%, bars the areerror filled bars with are red.filled with red. AccordingAccording to to Figure Figure 1010,, the error bars of the data are all filledfilled with green. This proves thatthat the the difference difference between between the the observed observed water water level level and and the the calculated calculated water water level level of of the the modelmodel is is within within the the confidence confidence interval interval and and that that the the groundwater groundwater flow flow model model is is reliable. reliable. ThisThis means means that that the the model model can can truly truly reflect reflect the the changes changes in in the the groundwater groundwater flow flow field field in in thethe study study area area and and can can be be directly directly used used in in the the ammonia ammonia migration migration model. model.

FigureFigure 10. 10. FittingFitting effect effect of of the the simulated simulated water water level level and and the the measured measured water water level. level.

2.3.2.3. Model Model of of Migration Migration of of Pollutants Pollutants in in the the Groundwater Groundwater OnOn the the basis basis of of the the above above model, model, the the model model of of migration migration of of pollutants pollutants in in the the ground- ground- waterwater can be establishedestablished byby usingusing MT3DMS MT3DMS of of GMS. GMS. The The partial partial differential differential equation equation of theof themodel model of the of convection-diffusionthe convection-diffusion of pollutants of pollutants in saturated-unsaturated in saturated-unsaturated soils is soils as follows: is as follows:  ∂(θC) ∂ ∂C ∂  = θ − (θ ) + +  Dij viC qsCs ∑ Rn ∂θC ∂  ∂C∂u ∂x∂i ∂xi ∂xi ⎧ = θD − θv C +q C +R = ∈ = (6) ∂u ∂x  C(∂xx, y, z, t) ∂xc 0(x, y, z) Ω, t 0  (6)  ⎨ Cx, y, z, t =cCx,(x, y, y, z z, ∈ t) Ω,= c ( x, y, z ) ∈ Γ 1 , t = 0 t ≥ 0 ⎩Cx, y, z, t =cx, y, z ∈Г, t ≥ 0 where each part of the equation is as follows: whereC—concentration each part of ofthe the equation dissolved is as phase follows: of the pollutants (ML−3); C—concentrationθ—porosity of the of stratum the dissolved medium phase (dimensionless); of the pollutants (ML−3); t—time (T); xi—distance along the axis of the rectangular coordinate system (L); 2 −1 Dij—tensor of the coefficient of hydrodynamic dispersion (L T ); −1 vi—actual flow rate of pore water on average (LT ); qs—flow of the aquifer per unit volume, representing source (positive) and sink (negative) (T−1); Water 2021, 13, x FOR PEER REVIEW 11 of 17

θ—porosity of the stratum medium (dimensionless); t—time (T);

x—distance along the axis of the rectangular coordinate system (L); 2 −1 Water 2021, 13, 1924 D—tensor of the coefficient of hydrodynamic dispersion (L T ); 11 of 17 −1 v—actual flow rate of pore water on average (LT ); q—flow of the aquifer per unit volume, representing source (positive) and sink (negative) (T−1); −3 C —concentration of pollutants in a source or sink (ML−3 ); Cs—concentration of pollutants in a source or sink (ML ); − − −3 3−1 1 ∑∑RRn——termterm of of chemical chemical reaction reaction (ML (ML TT); and); and ΩΩ—overall—overall simulation area.

2.3.1.2.3.1. SelectionSelection ofof SimulatedSimulated FactorsFactors AccordingAccording toto thethe hydraulichydraulic gradientgradient ofof thethe groundwatergroundwater alongalong thethe FengFeng River,River, fourfour riverriver profilesprofiles perpendicularperpendicular toto thethe flowflow directiondirection ofof thethe groundwatergroundwater werewere selectedselected inin thethe studystudy areaarea andand divideddivided thethe riverriver intointo fivefive sections.sections. TheThe fourfour profilesprofiles areare A,A, B,B, C,C, andand DD fromfrom southsouth toto northnorth (Figure(Figure 11 11).).

FigureFigure 11.11. PositionsPositions ofof thethe fourfour proﬁles.profiles.

EachEach proﬁleprofile waswas sampled,sampled, thethe concentrationconcentration ofof ammoniaammonia inin thethe riverriver waswas measured,measured, andand thethe concentrationconcentration was was generalized generalized as as the the input input concentration concentration at theat the upper upper boundary boundary of theof the ammonia ammonia migration migration model model (Table (Table3). 3).

TableTable 3. 3.Concentrations Concentrations ofof ammoniaammonia in in each each proﬁle. profile.

Proﬁle NumberProfile Number A B A B C C D D ConcentrationConcentration (mg/L) (mg/L) 0.48 0.46 0.48 0.46 0.34 0.34 0.410.41

2.3.2. Stress Period 2.3.2. Stress Period The simulation started from 1 January 2021 to 31 December 2040. Each corresponding The simulation started from 1 January 2021 to 31 December 2040. Each corresponding calendar month was taken as a stress period, and then, the whole simulation period was calendar month was taken as a stress period, and then, the whole simulation period was divided into 240 stress periods. divided into 240 stress periods.

3.3. ResultsResults 3.1.3.1. TheThe AreaArea ofof InﬂuenceInfluence ofof AmmoniaAmmonia TheThe area of influence inﬂuence of of ammonia ammonia from from th thee Feng Feng River River on the on thegroundwater groundwater was wassim- simulatedulated (Figure (Figure 12). 12 ). WaterWater2021 2021, 13, 13, 1924, x FOR PEER REVIEW 1212 ofof 17 17

(a) (b)

FigureFigure 12. 12.Simulation Simulation results results of of plane plane migration migration of of the the ammonia ammonia pollution pollution plume. plume.( (aa)) Simulation Simulation results results of of plane plane migration migration ofof the the ammonia ammonia pollution pollution plume plume for for 100 100 days days (b )(b Simulation) Simulation results results of planeof plane migration migration of the of ammoniathe ammonia pollution pollution plume plume for 1000for 1000 days days (c) Simulation (c) Simulation results results of plane of plane migration migration of the of ammoniathe ammonia pollution pollution plume plume for 3600 for 3600 days days (d) Simulation(d) Simulation results results of planeof plane migration migration of the of ammoniathe ammonia pollution pollution plume plume for 7300for 7300 days. days.

3.2.3.2. The The Ammonia Ammonia Pollution Pollution Plume Plume TheThe center center of of the the ammonia ammonia pollution pollution plume plume gradually gradually moves moves eastward eastward with with the the ﬂow flow ofof groundwater groundwater over over time. time. TheThe migrationmigration ofof ammoniaammonia isis similarsimilar toto that that of of the the center center of of ammoniaammonia pollution pollution plume, plume, and and ammonia ammonia accumulates accumulates in thein the lower lower part part of theof the river. river. During Dur- theing simulation the simulation period, period, the ammonia the ammonia pollution pollution plume plume gradually gradually becomes becomes wider wider and theand diffusionthe diffusion range range of ammonia of ammonia in the in groundwater the groundwater increases increases (Figure (Figure 13). 13).

Water 2021, 13, 1924 13 of 17 Water 2021, 13, x FOR PEER REVIEW 13 of 17

section A

section B

section C

section D

(a)

section A

section B

section C

section D

(b)

section A

section B

section C

section D

(c)

section A

section B

section C

section D

(d)

Figure 13. Simulation results of migration of the ammonia pollution plume in the four sections. (a) Simulation results of plane migration of ammonia pollution (b) Simulation results of plane migration of ammonia pollution (c) Simulation results of plane migration of ammonia pollution (d) Simulation results of plane migration of ammonia pollution. Water 2021, 13, x FOR PEER REVIEW 14 of 17

Water 2021, 13, 1924 14 of 17 Figure 13. Simulation results of migration of the ammonia pollution plume in the four sections.(a) Simulation results of plane migration of ammonia pollution (b) Simulation results of plane migration of ammonia pollution (c) Simulation results of plane migration of ammonia pollution (d) Simulation results of plane migration of ammonia pollution. 3.3. The Ammonia Content 3.3. The Ammonia Content Well A at about 50 m from the river way and well B at about 200 m from the river way wereWell taken A asat theabout objects 50 m of from observation the river to way observe and thewell changes B at about in ammonia 200 m from content the river near waywells were during taken the as simulation the objects period of observation (Figure 14 to). observe the changes in ammonia content near wells during the simulation period (Figure 14).

(a)

(b)

Figure 14. Changes in the ammonia content near wells. ( (aa)) Changes Changes in in ammonia ammonia content content near near well well A. ( b) Changes in ammonia content near well B.

The ammonia content content near near the the wells close to the watercourse is greatly affected by thethe river. river. At At the the beginning beginning of of the the simulation, simulation, ammonia ammonia in in the the river river migrates migrates to to the the wells. wells. The ammonia content near well A stabilizes at about 0.30 mg/Lmg/L around around the the year year 2040, while the the ammonia ammonia content content near near well well B stabilized B stabilized at about at about 0.06 0.06 mg/L mg/L around around the year the 2031. year In2031. general, In general, the influence the inﬂuence of ammonia of ammonia from fromthe river the river on the on groundwater the groundwater is within is within an ac- an ceptableacceptable range, range, and and the the ammonia ammonia content content near near the the wells wells meets meets the the Quality Quality Standard Standard for Ground Water IIIІІІ [ammonia (GB/T14848-2017 (in (in Chinese)]. Chinese)]. 4. Discussion 4. Discussion The reasons for the above situation are as follows: The reasons for the above situation are as follows: In a water body, ammonia is an important parameter indicative of inorganic pollution In a water body, ammonia is an important parameter indicative of inorganic pollu- that mainly result from anthropogenic impacts [34,35]. It was used as the primary restrictive tion that mainly result from anthropogenic impacts [34,35]. It was used as the primary descriptor for the water pollution control goals in the 12th Five-Year Plan of China (2011– 2015) [36]. The reference ammonia concentration value established by the Quality Standard for Ground Water class III (GB/T14848-2017 (in Chinese)) is 0.5 mg/L. The sampling sites of this study are all located 5–12 m downstream of the village’s sewage outfall in urban Water 2021, 13, 1924 15 of 17

areas and are presumably influenced by industrial effluent or domestic sewage discharge, which may contribute to the increased ammonia levels at these points. Apart from that, poultry, livestock, and agricultural fertilizers also attribute to the increase in ammonia. In a town in Kunshan, China [37], the migration of groundwater pollution gradually spreads from the central area to the surrounding area, and the research center has the high- est concentration and gradually decreases along the diffusion direction. This is because the surface elevation of the river valley is lower than that of other areas, and the groundwater level here is lower than that of other areas as well. Meanwhile, the contour lines under the river valley are fewer and the groundwater flows relatively slower. Therefore, as ammonia in the river migrates to groundwater under the river way, it accumulates. Different from another study, this study detected that, due to the pumping of groundwater at 50,000 t/d since 1958, when Xi’an the 3rd waterworks established the east Feng River bank, the groundwater level on the east bank of the Feng River dropped and a regional depression cone formed around the wells. Then, the ammonia in the river seeps into the groundwater and moves toward the east bank of the Feng River with the groundwater flow. Finally, it accumulates around the depression cone of the groundwater on the east bank of the Feng River, leading to an increase in the concentration of ammonia in the groundwater on the east bank of the Feng River, with the main impact range within 200 m along the Feng River. Apart from domestic pollutants from residents along the river, waste water, and the industrial “three wastes” resulting in ammonia concentration in the river, this study finds that nitrogen fertilizers and sewage irrigation also have varying degrees of impact on the ammonia content of groundwater since the study area is located among farmland. There- fore, the ammonia content of the groundwater sampling points far from the watercourse is higher than that closer to the watercourse.

5. Conclusions (1) During this simulation, a field survey was conducted in the study area. Combined with the current hydrology and water quality of the study area, the hydrogeological conceptual model was converted into a numerical model of groundwater flow and solute transport. Additionally, by using ammonia as the simulation factor, the cali- brated model was used to simulate the hydrology and water quality of the study area. It is concluded that the established hydrogeological conceptual model and numerical model are correct, and the selected parameters and calculated source and sink terms are reasonable, which conform to the actual groundwater conditions in the study area. Thus, the results of this study can be used for water flow field research and water source mining planning. (2) In this study, the GMS model was used to predict and analyze the diffusion process of pollutants in the study area and to simulate the migration trend for 20 years. Through the analysis of the GMS model, it is predicted that pollutants will gradually migrate eastward from the river to groundwater, with the impact of ammonia in the river water on groundwater remaining within an acceptable range, and the ammonia content at the intake well meets the Quality Standard for Ground Water class III [ammonia (GB/T14848-2017 (in Chinese))]. (3) Using the GMS model to simulate and predict the hydrological and water quality status of groundwater is conducive to understanding the status of groundwater pollution and to adopting thoughtful treatment and maintenance measures. However, the speed of pollutant migration and the size of the diffusion range are closely related to the amount and intensity of groundwater extraction. Therefore, when predicting the migration of pollutants, the influencing factors should be taken into consideration.

Author Contributions: Formal analysis, methodology, software, visualization, writing—original draft, writing—review and editing, Z.W.; funding acquisition, methodology, writing—review and editing, X.Z.; writing—review and editing, T.X.; data curation, N.W.; investigation, J.Y. All authors have read and agreed to the published version of the manuscript. Water 2021, 13, 1924 16 of 17

Funding: This research was funded by China postdoctoral science foundation, grant number 2018M643689. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Part of the data (observations/measurements) used in this paper was gotten from field study or experiments conducted by my team. The other part of data was taken after application from the third waterworks in Xi’an and we are responsible for these data. The results of this study are freely available. Acknowledgments: The third waterworks in Xi’an, China, provided the pumping wells and water quality data. Conflicts of Interest: The authors declare no conflict of interest.

References 1. Mercado, A.; Libhaber, M.; Soares, M.I.M. In situ Biological Groundwater Denitrification: Concepts and Preliminary Field Tests. Water Sci. Technol. 1988, 20, 197–209. [CrossRef] 2. Darabi, H.; Torabi Haghighi, A.; Mohamadi, M.; Rashidpour, M.; Ziegler, A.; Hekmatzadeh, A.A.; Klöve, B. Urban flood risk mapping using data-driven geospatial techniques for a flood-prone case area in Iran. Hydrol. Res. 2019, 51, 127–142. [CrossRef] 3. Junjie, J. A Study of the Grain Production and Fertilizer Application in China-From the Perspective of Peasant Households and Market. Master’s Thesis, Shandong University, Jinan, China, 2016. 4. Arnold, J.; Moriasi, D.; Gassman, P.; Mikayilov, F.; White, M.; Srinivasan, R.; Santhi, C.; Harmel, R.; van Griensven, A.; Van Liew, M.; et al. SWAT: Model use, calibration, and validation. Trans. ASABE 2012, 55, 1491–1508. [CrossRef] 5. Beaujouan, V.; Durand, P.; Ruiz, L.; Aurousseau, P.; Cotteret, G. A hydrological model dedicated to topography-based simulation of nitrogen transfer and transformation: Rationale and application to the geomorphology—Denitrification relationship. Hydrol. Process. 2002, 16, 493–507. [CrossRef] 6. He, J.; Li, A.; Tan, Z.; Xie, G.; Shi, H.; Yao, J.; Zhang, X.; Sun, W.; Du, P. Potential impact of water transfer policy implementation on lake eutrophication on the Shandong Peninsula: A difference-in-differences approach. Hydrol. Res. 2020, 51, 1063–1076. [CrossRef] 7. Li, X.; Xu, Y.; Li, M.; Ji, R.; Dolf, R.; Gu, X. Water Quality Analysis of the Yangtze and the Rhine River: A Comparative Study Based on Monitoring Data from 2007 to 2018. Bull. Environ. Contam. Toxicol. 2021, 106, 825–831. [CrossRef] 8. Yuepeng, L. Research on the Influence of Polluted River on Groundwater Quality and the Model of Protection and Rehabilitation. Ph.D. Thesis, Chang’an University, Xi’an, China, 2017. 9. Rankinen, K.; Granlund, K.; Etheridge, R.; Seuri, P. Valuation of nitrogen retention as an ecosystem service on a catchment scale. Hydrol. Res. 2013, 45, 411–424. [CrossRef] 10. De la Hoz, R.E.; Schlueter, D.P.; Rom, W.N. Chronic lung disease secondary to ammonia inhalation injury: A report on three cases. Am. J. Ind. Med. 1996, 29, 209–214. [CrossRef] 11. Casal, L.; Durand, P.; Akkal-Corfini, N.; Benhamou, C.; Laurent, F.; Salmon-Monviola, J.; Ferrant, S.; Probst, A.; Probst, J.-L.; Vertès, F. Reduction of stream nitrate concentrations by land management in contrasted landscapes. Nutr. Cycl. Agroecosystems 2019, 114, 1–17. [CrossRef] 12. Yao, J.; Wang, G.; Xue, B.; Xie, G.; Peng, Y. Identification of regional water security issues in China, using a novel water security comprehensive evaluation model. Hydrol. Res. 2020, 51, 854–866. [CrossRef] 13. Ferrant, S.; Oehler, F.; Durand, P.; Ruiz, L.; Salmon-Monviola, J.; Justes, E.; Dugast, P.; Probst, A.; Probst, J.-L.; Sanchez-Perez, J.-M. Understanding nitrogen transfer dynamics in a small agricultural catchment: Comparison of a distributed (TNT2) and a semi distributed (SWAT) modeling approaches. J. Hydrol. 2011, 406, 1–15. [CrossRef] 14. Lyra, A.; Loukas, A.; Sidiropoulos, P.; Tziatzios, G.; Mylopoulos, N. An Integrated Modeling System for the Evaluation of Water Resources in Coastal Agricultural Watersheds: Application in Almyros Basin, Thessaly, Greece. Water 2021, 13, 268. [CrossRef] 15. Evenson, G.R.; Golden, H.E.; Lane, C.R.; D’Amico, E. An improved representation of geographically isolated wetlands in a watershed-scale hydrologic model. Hydrol. Process. 2016, 30, 4168–4184. [CrossRef] 16. Farjad, B.; Gupta, A.; Razavi, S.; Faramarzi, M.; Marceau, D.J. An Integrated Modelling System to Predict Hydrological Processes under Climate and Land-Use/Cover Change Scenarios. Water 2017, 9, 23. [CrossRef] 17. Mabon, F.; Raimbault, T.; Moreau, P.; Devienne, S.; Delaby, L.; Durand, P.; Ruiz, L.; Vertes, F. Concilier efficacité technico- économique et environnementale des exploitations agricoles en zone vulnérable : Apport du diagnostic agraire. Fourrages 2009, 199, 373–388. 18. Petry, J.; Soulsby, C.; Malcolm, I.A.; Youngson, A.F. Hydrological controls on nutrient concentrations and fluxes in agricultural catchments. Sci. Total Environ. 2002, 294, 95–110. [CrossRef] Water 2021, 13, 1924 17 of 17

19. Ohte, N.; Tokuchi, N.; Katsuyama, M.; Hobara, S.; Asano, Y.; Koba, K. Episodic increases in nitrate concentrations in streamwater due to the partial dieback of a pine forest in Japan: Runoff generation processes control seasonality. Hydrol. Process. 2003, 17, 237–249. [CrossRef] 20. Lapworth, D.J.; Shand, P.; Abesser, C.; Darling, W.G.; Haria, A.H.; Evans, C.D.; Reynolds, B. Groundwater nitrogen composition and transformation within a moorland catchment, mid-Wales. Sci. Total Environ. 2008, 390, 241–254. [CrossRef][PubMed] 21. Schubert, J. Hydraulic aspects of riverbank filtration—Field studies. J. Hydrol. 2002, 266, 145–161. [CrossRef] 22. Balke, K.-D.; Zhu, Y. Natural water purification and water management by artificial groundwater recharge. J. Zhejiang Univ. Sci. B 2008, 9, 221–226. [CrossRef][PubMed] 23. Alvarez-Cobelas, M.; Angeler, D.G.; Sánchez-Carrillo, S. Export of nitrogen from catchments: A worldwide analysis. Environ. Pollut. 2008, 156, 261–269. [CrossRef][PubMed] 24. Johnsson, H.; Bergstrom, L.; Jansson, P.-E.; Paustian, K. Simulated nitrogen dynamics and losses in a layered agricultural soil. Agric. Ecosyst. Environ. 1987, 18, 333–356. [CrossRef] 25. Vanclooster, M.; Viaene, P.; Diels, J.; Feyen, J. A deterministic evaluation analysis applied to an integrated soil-crop model. Ecol. Model. 1995, 81, 183–195. [CrossRef] 26. Jabro, J.D.; Toth, J.D.; Dou, Z.; Fox, R.H.; Fritton, D.D. Evaluation of Nitrogen Version of Leachm for Predicting Nitrate Leaching. Soil Sci. 1995, 160, 209–217. [CrossRef] 27. Knisel, W.G. CREAMS: A Field-Scale Model for Chemicals, Runoff and Erosion from Agricultural Management Systems; U.S. Department of Agriculture, Science and Education Administration: Washington, DC, USA, 1980; p. 640. 28. Medici, C.; Bernal, S.; Butturini, A.; Sabater, F.; Martin, M.; Wade, A.J.; Frances, F. Modelling the inorganic nitrogen behaviour in a small Mediterranean forested catchment, Fuirosos (Catalonia). Hydrol. Earth Syst. Sci. 2010, 14, 223–237. [CrossRef] 29. Moharir, K.; Pande, C.; Patil, S. Inverse modelling of aquifer parameters in basaltic rock with the help of pumping test method using MODFLOW software. Geosci. Front. 2017, 8, 1385–1395. [CrossRef] 30. Wang, R.; Liu, Z.; Yao, Z.; Lei, Y. Modeling the risk of nitrate leaching and nitrate runoff loss from intensive farmland in the Baiyangdian Basin of the North China Plain. Environ. Earth Sci. 2014, 72, 3143–3157. [CrossRef] 31. Yousefi, H.; Zahedi, S.; Niksokhan, M.H.; Momeni, M. Ten-year prediction of groundwater level in Karaj plain (Iran) using MODFLOW2005-NWT in MATLAB. Environ. Earth Sci. 2019, 78, 343. [CrossRef] 32. Rankinen, K.; Lehtonen, H.; Granlund, K. Assessing the Effects of Agricultural Change on Nitrogen Fluxes Using the Integrated Nitrogen CAtchment. In Proceedings of the 2nd International Congress on Environmental Modelling and Software, Osnabrück, Germany, 14–17 June 2004. 33. Ferrant, S.; Durand, P.; Justes, E.; Probst, J.-L.; Sanchez-Perez, J.-M. Simulating the long term impact of nitrate mitigation scenarios in a pilot study basin. Agric. Water Manag. 2013, 124, 85–96. [CrossRef] 34. Bu, J.; Sun, Z.; Ma, R.; Liu, Y.; Gong, X.; Pan, Z.; Wei, W. Shallow Groundwater Quality and Its Controlling Factors in the Su-Xi-Chang Region, Eastern China. Int. J. Environ. Res. Public Health 2020, 17, 1267. [CrossRef][PubMed] 35. El Idrysy, H.; De Smedt, F. Modelling groundwater flow of the Trifa aquifer, Morocco. Hydrogeol. J. 2006, 14, 1265–1276. [CrossRef] 36. Sun, R.; Wang, Z.Z.; Chen, L.; Wang, W.W. Assessment of Surface Water Quality at Large Watershed Scale: Land-Use, Anthro- pogenic, and Administrative Impacts. J. Am. Water Resour. Assoc. 2013, 49, 741–752. [CrossRef] 37. Junyan, Z.; Cuimei, L.; Jingxiong, H.; Song, X. Application of GMS Model in Hydrological and Water Quality Simulation. J. China Hydrol. 2019, 39, 56–73.