Introduction

The sources of water pollution have recently increased and have become more diverse (Kampbell et al. 2003; Nolan and Stoner 2000; Jacobs 1985), thus exacerbating the underground water pollution problem. Nitrates are typical pollutants in underground water systems and have become the focus of international attention. Nitrates can be both point and non-point sources of pollution because they can originate from agrochemicals (Harter et al. 2002), wastewater irrigation (Reddy 1984), or the intensive feeding of domestic animals (Hallberg 1993). The nitrate content in underground water resources is currently increasing. Agriculture is the main source of nitrate nitrogen pollution in groundwater (Gonzalez-Herrera et al. 2014; Edet 2014; Huang et al. 2013; Klammler et al. 2013). Furthermore, nitrates are highly water soluble, have weak adsorption, and exhibit strong mobility. Nitrates readily leach through vertical migration via atmospheric precipitation or irrigation water. This leaching leads to catchment pollution via lateral recharge and eventually reaches the groundwater–soil medium. If the vadose zone is particularly thin, the pollution of groundwater is likely (Gonzalez-Herrera et al. 2014; Rao et al. 2013). The vertical recharge caused by atmospheric precipitation and the lateral recharge caused by rivers can both lead to variations in the groundwater level as well as those in soil–water indicators (Ashworth and Shaw 2006; Ashworth et al. 2008). In turn, these fluctuations significantly affect the migration and transformation of underground water pollutants. Hence, a systematic understanding of the variations in nitrate content caused by underground water table fluctuation is vital in environmental research.

In recent years, most studies on water table variations have focused on the light non-aqueous phase liquid (LNAPL)/dense non-aqueous phase liquid migration law in porous media, the saturation–capillary pressure hysteresis effect caused by water table fluctuation, and the numerical simulation of variations in the bottom line of unsaturated zones (Steffy et al. 1998; Kampbell et al. 2003). For instance, Tanner et al. (1999) determined that water table fluctuation in a constructed wetland system is conducive to total nitrogen (TN) and NH4 +–N removal and that the removal efficiency is closely associated with vibration frequency. Indoor-soil column research by Rainwater et al. (1993) showed that water table fluctuation can accelerate diesel degradation in soil. Kamon et al. (2007) determined the transport properties of LNAPL in porous media and conducted LNAPL migration research on the effects of water table fluctuation. However, laboratory simulations can reflect natural processes to a certain extent but cannot demonstrate the complexities present in the actual field. Consideration of impact factors to ensure accuracy can be difficult. Moreover, the various degrees of error inherent in simulations increase the difficulty of translating the results into actual, precise laws. Therefore, specific reports on the effects of water table variations on the variation trend of nitrate content in actual underwater systems are rare.

Thus, this study uses a case study analysis and collects currently available materials to investigate the soil–water physicochemical properties as well as the variation trend of nitrate content under the effects of water table fluctuation in five monitoring points in the Luoyang area, China.

Site context

Physiographic conditions

The research area is located in Luoyang (112°20′ to 112°35′E, 34°35′ to 34°47′N) within Yuxi Luoyang Basin, the location shown in Fig. 1. This area is located north of Mang Mountain, south of Yique, west of Small Qinling Basin, and east of Yanshi Plain. The locality is in the transition area between the southern edge of a warm temperate zone and the northern edge of a subtropical transition zone. The annual average surface air temperature is approximately 15 °C, with a maximum temperature of 40.4 °C and a minimum temperature of 20.2 °C. The annual average precipitation is approximately 630 mm.

Fig. 1
figure 1

Location map of study area

Full size image

Hydrogeological conditions

Groundwater in the study area is mainly loose rock pore water, and the aquifer thickness is greater than 50 m. The groundwater occurs mainly in the plain area of Luoyang and is stored in Cenozoic Quaternary sediments. Overall, the groundwater flows from west to east, and the aquifer permeability coefficient is between 20 and 50 m/d. Important recharge and water sources of the groundwater include rainfall and the Yiluo River. In the study area, the Tertiary unit includes abundant loose sediment, creating favorable conditions for groundwater recharge, runoff, and storage. Thus, shallow groundwater resources are highly abundant.

Monitoring point information

Basic information

Figure 2 shows the five typical points selected for analysis of the effects of water table fluctuation on nitrate movement based on survey data. Of these, JC1, JC3, and JC5 are wells. JC1 and JC3 are both drinking wells and production wells, JC5 is a drinking well, and JC2 and JC4 are both hydrological observation holes. The depths of JC1, JC2, JC3, JC4, and JC5 monitoring points are 61, 83, 75, 85, and 70 m, respectively. All monitoring points are drilled in drinking water sources.

Fig. 2
figure 2

Map showing the monitoring points in the survey area

Full size image

Geological information

The five monitoring points were selected from various areas. Figure 3 shows significant differences in stratum structures, which indicate large discrepancies in the permeability of the sola and considerable complexity in the soil environment. Groundwater indicators tested at these points between 2007 and 2011 were used to discuss the effects of water table fluctuation on nitrate content.

Fig. 3
figure 3figure 3figure 3

Borehole diagrams of the five monitoring points

Full size image

Human activities in the sites and surrounding areas

In recent years, varying degrees of nitrate nitrogen, ammonium nitrogen, and nitrite nitrogen have been detected and analyzed in the five selected areas. Because agricultural and industrial activities are the main sources of groundwater contamination (Gonzalez-Herrera et al. 2014; Huang et al. 2013; Klammler et al. 2013), we investigated such activities occurring within 1 km of each point. These agricultural and industrial activities include irrigation, chemical fertilizer use, sewage discharge (Wang et al. 2013), and massive use of groundwater. Because such activities can hasten nitrate seepage into the aquiclude across unsaturated zones, we selected them as survey indicators. The investigation results are shown in Table 1.

Table 1 Investigation of the areas surrounding the monitoring points
Full size table

Methods of sampling and detection

The type of water sampling in which pumps were used was instantaneous. We allowed the water to flow for several minutes to discharge the original water in the pipe; thereafter, the water samples were stored in 200 ml glass containers. We then placed each sample in a portable sample storage box. The samples were sent to the laboratory soon after sampling was completed. Nitrates were detected using the water quality determination method through nitrate–nitrogen ultraviolet spectrophotometry. Ammonia was detected by nitrogen–Nessler’s reagent spectrophotometry. Nitrite nitrogen was detected by the nitrogen (nitrite)-spectrophotometric method.

Methodology

In this study, preliminary work focused on the collection of information and data in terms of hydrogeology such as landforms, landscape, and geology and groundwater dynamic indicators such as water temperature, pH, dissolved oxygen (DO), redox potential (E h), nitrate nitrogen concentration, and ammonium nitrogen concentration. Field data were collected during a five-year period beginning in January 2007. Water table variations consistent with the collected data were analyzed quarterly. In the experiments, water was collected by a Baylor tube, and the water table was monitored by a water level monitoring instrument. Wellhead coordinates and elevation were calculated by a Global Positioning System. Land use within 1 km of the monitoring points was surveyed according to underground water use, nitrogen fertilizer use, factory discharge, and animal husbandry. The survey forms were completed and the data were organized on the basis of local natural geography, geology, and hydrogeologic conditions. The results were then mapped and statistically analyzed by Microsoft Excel 2007, Origin 8.0, and SPSS 17.0.

Results and discussion

Water table mobility

Seasonal precipitation is a major contributor to regional variations in the water table. The rainy season generally occurs from July to September in the study area and accounts for 70 % of the annual precipitation. Moreover, agricultural land in JC1, JC3, and JC5 is pump-irrigated in April, May, and November each year, and the irrigation frequency and pump output are inversely related to local precipitation. Underground water is used as a production material by major industries near JC1 and JC3. Thus, the effects of water table variations in these areas are complicated and are difficult to understand. However, the most important factors affecting water table fluctuation in such areas are precipitation and irrigation water. The variations in groundwater depth at the five monitoring points are shown in Fig. 4. Statistics for the water table are shown in Table 2.

Fig. 4
figure 4

Dynamic graph of water levels at the five monitoring points from 2007 to 2011

Full size image
Table 2 Water level characteristics of the monitoring points
Full size table

Variations in the dissolved oxygen content

Figure 5 shows variations in the dissolved oxygen (DO) concentration at the monitoring points. A significant variation was observed at each point. The DO concentration peaked in July, with DO values higher than those in other months. The DO concentration in JC5 was higher than those at other points. The dynamic characteristics of underwater DO concentration at each point are given in Table 3.

Fig. 5
figure 5

Variations in the dissolved oxygen (DO) content in groundwater at the five monitoring points

Full size image
Table 3 Characteristics of groundwater pH at each monitoring point
Full size table

Variations in pH and E h

The pH levels at the monitoring points showed periodic variations, although the variations were not significant. The five-year pH-variation statistics are shown in Table 4.

Table 4 Characteristics of groundwater pH at each monitoring point
Full size table

Table 5 shows the five-year statistical characteristics of E h. Figure 6 shows that E h is inversely correlated with the variation in the water table. For example, Fig. 4a shows that the data of point JC1 indicate a decrease in groundwater depth from 8.5 to 9.3 m from October 2007 to April 2008, whereas E h increased from 157 to 219 mV. The groundwater depth increased from 9.3 to 8.55 m during the subsequent 3 months, from April to July 2008, whereas E h decreased from 219 to 176 mV. In the succeeding season, from July 2008 to January 2009, the groundwater depth decreased from 8.55 m to 9.5 m, whereas E h increased to 210 mV.

Table 5 Characteristics of groundwater redox potential (E h) at each monitoring point
Full size table
Fig. 6
figure 6

Variations in groundwater pH and redox potential (E h) at the various points from 2007 to 2011

Full size image

E h has a higher possibility of showing uniformity with groundwater depth over long-term seasonal variations. Increases in groundwater depth generally cause a decrease in E h and vice versa.

Temperature variations

Figure 7 shows a significant correlation between water temperature and seasonal temperature. For example, the winter temperature of water at JC4 was at minimum between December and February, whereas the peak temperature appeared annually from June to August. Table 6 shows that the difference in temperature at various groundwater depths was not significant and that the average temperature generally ranged from 16.15 to 18.2 °C.

Fig. 7
figure 7

Variations in groundwater temperature at each monitoring point from 2007 to 2011

Full size image
Table 6 Groundwater temperature trends at each monitoring point
Full size table

Variations in the nitrate nitrogen and ammonium nitrogen contents

Figure 8 shows the time-dependent variations in groundwater nitrate nitrogen and ammonium nitrogen concentrations in the five monitoring points, whereas Table 7 shows the overall characteristics. The variations in nitrate nitrogen and ammonium nitrogen contents in JC1 were significant and negatively correlated. The five-year statistics of nitrite nitrogen content show that this indicator has an average value of 0.002 ± 0.001 mg L−1 and does not follow a significant law. This trend may be attributed to the instability of the nitrification and denitrification intermediates; NO2 − can be readily oxidized.

Fig. 8
figure 8

Variations in nitrate nitrogen and ammonium nitrogen concentrations at the different monitoring points

Full size image
Table 7 Characteristics of nitrate nitrogen and ammonium nitrogen concentrations at each monitoring point
Full size table

Relevance of the groundwater dynamic index and water table fluctuation

From “Monitoring point information”, we know the use function and the surrounding environment are relatively similar among JC1, JC3, and JC5. Moreover, these parameters are similar between JC2 and JC4. Therefore, two typical areas, JC1 and JC2, are analyzed in this section. The correlations and significance of groundwater environment indicators and water table fluctuation data were analyzed by SPSS 17.0 on the basis of the hydrogeologic conditions of the sites. The variations in the impact factors such as water temperature, pH, E h, water table variation and concentrations of nitrate nitrogen, ammonium nitrogen, and nitrite nitrogen at the various points were analytically monitored from 2007 to 2011. The water temperature, pH, and E h were found to be directly and separately affected by variations in the water table. The effects of water table fluctuation on the contents of nitrate nitrogen, ammonium nitrogen, and nitrite nitrogen were also analyzed by considering the hydrogeologic conditions of the sites.

The correlation and significance of the various parameters, particularly the JC1 and JC2 groundwater indicators, were determined by SPSS 17.0 (Tables 8, 9).

Table 8 Correlation of groundwater parameters in JC1
Full size table
Table 9 Correlation of groundwater parameters in JC2
Full size table

Table 8 shows a significant correlation of groundwater depth and E h and nitrate nitrogen. The nitrate nitrogen and ammonium nitrogen concentrations were significantly and negatively correlated. A negative correlation was also observed between E h and ammonium nitrogen content and between pH and nitrate nitrogen content. In a similar manner, the JC2 data in Table 9 indicate a significant positive correlation between the water-table and nitrate nitrogen content, whereas E h and ammonium nitrogen content were significantly and negatively correlated. A significant negative correlation was observed between nitrate nitrogen and ammonium nitrogen concentrations.

By JC1 and JC2 as examples in the investigation of water table fluctuation effects on different parameters, the following results were obtained:

Analysis of the relationship between E h and water table fluctuation

Significantly negative correlations were observed between E h and water table fluctuation in the two sites. Figure 9a shows the five-year seasonal variations in E h and water level in JC1 from January 2007 to October 2009 and from January 2011 to October 2011. Figure 9b shows the five-year seasonal variations in E h and water level in JC1 from January 2007 to April 2008, from January 2009 to January 2011, and from April 2011 to October 2011. Both figures show a negative correlation between E h and groundwater depth, possibly as a result of the increased water table. The distance between the water level and the ground surface decreased, which in turn increased the DO content E h. Otherwise, E h declined when the distance between the water level and the ground surface increased.

Fig. 9
figure 9

Variations in redox potential (E h) and water level in JC1 and JC2

Full size image

Analysis of the relationship between DO and water table fluctuation

Theoretically, the distance between groundwater and the ground surface is small, and air can readily reach the groundwater. However, the current results do not indicate a significant correlation between groundwater DO and groundwater depth in the research areas, possibly because of the complexity of the hydrogeological conditions of the sites and utilization of the surrounding land.

Analysis of the effects of water table fluctuation on nitrate nitrogen and ammonium nitrogen concentrations

According to Tables 8 and 9, a significantly negative correlation exists between water table fluctuation and ammonium nitrogen concentration as well as that between the concentrations of nitrate nitrogen and ammonium nitrogen. According to the analyses discussed in previous sections, the variation in pH with water table fluctuation is not significant when the effects of precipitation on water table fluctuation are ignored. Moreover, the pH varies in the two sites ranged from 6.6 to 8.5 and had negligible effects on microbial activity. Therefore, pH variation is not considered as a major contributor to the variations in nitrate nitrogen and ammonium nitrogen concentrations. The increased water table caused the groundwater DO and E h to increase, which in turn promoted microbial activity during soil nitrification. Consequently, nitrification was increased but denitrification was inhibited until the nitrate nitrogen concentration increased and the ammonium nitrogen content decreased. Moreover, the high adsorbability and effumability of ammonium nitrogen resulted in its rapid absorption by soil as the water table increased or the groundwater depth decreased. In addition, closer proximity to the surface indicated high volatility, which is among the primary reasons for the decrease in ammonium nitrogen concentration in groundwater. Otherwise, the water table decreased and the groundwater depth increased, which in turn limited the contact between the air and groundwater. Consequently, DO was gradually consumed by microorganisms, and E h decreased as reducibility increased, which activated denitrifying bacteria. As a result, denitrification increased, whereas nitrification was inhibited. Eventually, the ammonium nitrogen concentration increased, but nitrate nitrogen was deoxidized to a lower state of nitrogenous compounds. These results clearly indicate that the effect of water table fluctuation on groundwater nitrate transformation is significant in field settings.

After several periodic water table fluctuations, the variations in concentrations of nitrate nitrogen, nitrite nitrogen, and ammonium nitrogen became insignificant compared with those under stable water-table conditions. This result may be attributed to the complex utilization of the areas surrounding the sites. Land-use practices such as the intensive use of chemical fertilizers and pesticides and the discharge of industrial wastewater can affect the nitrate content of groundwater.

Conclusion

Water table variations affect soil–water physicochemical properties such as DO and E h and therefore significantly affecting nitrate migration in soil sola. When the water table decreases, the nitrate nitrogen concentration increases, whereas the ammonium nitrogen concentration decreases because of the variations in physicochemical properties of the soil and underground water. Otherwise, the variations in concentration show opposite tendencies. The instability of nitrite nitrogen, which does not follow a specific law during the entire process, results in its low concentration. After the monitoring points experience several water table variations, the variations in the underwater index as well as those in the concentrations of nitrate nitrogen, ammonium nitrogen, and nitrite nitrogen become insignificant. The complexity of field conditions necessitates the consideration of all possible impact factors and indices. Therefore, water table variations should be considered in practical underwater impact assessments, pollutant migration simulations, and underwater blocking and repair.