Zhang Dayi¹, Li Guanghe¹, Guo Huaming¹, Lu Chang’ai², Zhangxu¹

1, Department of Environment Science and Engineering, Tsinghua University, Beijing, China, 100084,

2, China Agriculture Science Academy, Beijing, China,

Abstract: Local investigations and in-situ experiments were carried out in this research to investigate the nitrogen balance among water-soil-plant-air continuous (WSPAC) and the control factors of nitrogen balance and transport through unsaturated zones of soils in drainage area of Dianchi Lake. The results of local investigations showed that high contaminant of nitrate in groundwater was related to the degree and years of cultivation and 39% of the nitrogen which caused eutrophication of Dianchi Lake came from groundwater stream. In-situ experiments showed that a number of environmental factors influenced the balance of nitrogen in groundwater, five of which were fertilization, soil withhold, plant use, volatilization and Eluviate. The quantity of NO₃-N recharged into groundwater were related to fertilization and values of vertical fluxed into groundwater of NO₃-N were 19.30 (HF) and 5.52 (LF) g/plot. The vertical infiltration recharge was limited by groundwater level and hydraulic conductivity of soil. No significant difference of plant usage was found among treatments of 93.5-138.1 kg N/hm², while the percentage of nitrogen input was quite different, 20.8-30.4% in LF, higher than 8.5-11.5% in HF. Nitrogen volatilized into the air was related to the denitrification ability stimulated by organic matter supply and the ammonia volatilization relevant to cation exchange and evaporation in soil. Denitrification rate in unvegetated treatment was 6.36 mg/(kg·h), higher than 2.79 mg/(kg·h) in vegetated one. Vertical infiltration recharge was separately 0.363 mm/d in thick unsaturated zone while 1.851 mm/d on the other condition. The values of vertical fluxing into groundwater of NO₃-N were 19.30 g/plot and those of NO₂-N were 0.96 g/plot in thicker unsaturated zone, while 108.52 g/plot of NO₃-N and 0.11 g/plot NO2-N in thinner one, respectively.

Key words: nitrogen, unsaturated zones of soil, denitrification, balance and transport

1 Introduction

Groundwater is an important source of drinking water in the world and it is generally considered to be a very good source of drinking water because of its clarity and temperature stability. Recently, however, contamination of groundwater by pathogens such as Cryptosporidium and Giardia has been reported in the USA^[1]. Similar contamination can be expected to occur in China, and greater attention should, therefore, be given to preserve groundwater resources.

Contamination of groundwater by nitrate has occurred in many European countries and is also becoming a serious problem in China. In many cases, nitrate contamination in groundwater has become more and more serious recent years and it originates mainly from the excessive fertilization and uncontrolled land discharge of treated wastewater. Nitrate must be removed from drinking water because water containing a high concentration of nitrate can cause methemoglobinemia or blue baby syndrome in infants^[2]. In addition, an increased incidence of some forms of cancer from consumption of high-nitrate drinking water has been investigated, although no conclusions have yet been reached. The nitrate concentration in the water obtained from some groundwater wells in China has already reached 20mg N/L, which is the upper limit in current Chinese drinking water standards^[3]. The development of modern agriculture systems has resulted in high nitrate contaminants in groundwater in China, especially in drainage area of Dianchi Lake which is the most important planting base of vegetables and flowers and the contaminated groundwater that flowed into Dianchi Lake increased eutrophication.

2 Materials and methods

2.1 Experimental area

Dayu, as shown in figure 1 and located in approximately 25 Km southwest of Kunming city in southwestern China, lies on the east shore of Dianchi Lake, one of the largest lakes in Yunnan province. Mean temperature at this area is 14.7 ºC with a maximum of 31 ºC and minimum of -8.1 ºC. Mean annual precipitation is 782.5 mm, more than ninety percent of which falls from May to October.

Estimation of plant nutrient in Dianchi Lake drainage area showed that every year the nitrogen input into the soil by fertilizer is about 571.5´104t, which is 3.3 to 7.6 times higher than what the crop needs. Surplus of total nitrogen is 553.5´104t/a (about 711kg N/(hm·a)) including other nitrate resources, and these nitrogen resulted in damage to soil quality and pollution of disparated forms of water, especially groundwater, whose NO3-N quality concentration is up to 132.7 mg/L and 5.64 times higher above National Drinking Water Standard (20 mg/L).

2.2Experimental treatment

In situ experiments were conducted in the greenhouse cultivating celeries with an area of 46.5 × 4.0 m2 from February 25, 2003 to June 10, 2003. The pilot site occupied an area of 7.0 × 4.0 m2 within the greenhouse, which was divided into two plots, including HF (High Fertilization plot) and LF (Low Fertilization plot), with an area of 1.8 × 7.0 m2 for each plot. Fertilizations are summarized in Table 1. Irrigation intensity was same for all plots. The irrigation was generally carried out once every three days, with an essentially fixed value of 28.6 mm. Guo et al. (2004) described the irrigation intensity in more detailed.

Fugure 1 Villages’ Distribution in Dayu

2.3 Analytical methods

Prior to celery transplantation, TDR (TRIME-T3) and WM-1 tensiometers were installed at 0.3, 0.6, 1.0, 1.5, 2.0, 2.8 m BGL (Below ground level). TDR was used to monitor soil moistures with measured ranges of 0 and 60 % in volumetric moisture content and observed precision of 2%^[4，5], and WM-1 was used to monitor matrix potential^[6].

Simultaneously, the soil water samplers were set up at locations corresponding to TDR and WM-1. In order to avoid disturbing soil structure, the soils were carefully dug out and replaced at their original locations after installation of experimental equipments. A well, penetrating into 5 m BGL, was installed to monitor groundwater table to 2 m east of the experimental site during the experiment. Groundwater table slightly fluctuated during the study from a minimum of 151 cm, and a mean of 191 cm BGL.

Table 1 Fertilizations in the field experiment in Dayu, Kunming city

^a low fertilization plot; ^b high fertilization plot

Gas was collected by obturator box and gas velocity was controlled by flowmeter. NH3 was collected in 30min by 1 L/min and the adsorptive solution was H2SO4 (0.01N). NO/NO2 was collected in 30min by 0.2 L/min and the adsorptive solution was N-1-Naphthylenediumine dihydrochloride. N2O was detected by Shimadzu GC-14A gas chromatography meter. Column was filled with Porapak Q and carrying gas was N2 by 60 ml/min. Temperature of detect facility, column and entrance was 300, 60, 100℃ respectively. Standard N2O was provided by Nation Standard Material Research Centre and detected by CSIRO Air Institute, Australia.

Plant samples and soil samples were collected at the same time as follows: 20d, 40d, 60d and 90d.

3 Results and discussion

3.1 Nitrate pollution

The study area is completely covered by Quaternary loose deposits. Geologic form of the research area is mainly composed of Ⅳscarification, and hardpan and Q4a1 and Q4a1+1 is the main aquifers while the claypan is considered as the staunch. The upper of Q4a1 and Q4a1+1 is considered of floury sand with thickness of 0-10m and the lower of them is hardpan aquifer with thickness of 1.65-10m, whose structure is loose and well connectivity. This aquifer was mainly supplied by rainfall and surface water and drained into the Dianchi Lake from east to west.

3.1.1 Nitrate Plane Ddistribution

Nitrate polluted groundwater mainly distributed in seven areas as shown in figure2 which were Wulong village, Xin village, Tuluo village, Dahaiyan village, east farmland of Taipingguan, north farmland of Dayu and west farmland of Xiaohekou. Most of the NO3-N quality concentration was over 20 mg/l (National Drinking Water Standard) and up to 132.7 mg/L and 5.64 times higher above the Standard.

Fig. 2 Nitrate in groundwater distributing in research area

3.1.2 Nitrate Time Distribution

According to rainfall and evaporation data in twenty years, there is distinct rainy season and dry season. Rainy season is from May to October while the others are dry season.

Nitrate contents in groundwater in March and June were compared. In cultivation region, it was distinct difference that nitrate content was below measure threshold in dry season while 5-15 mg/L and up to 133 mg/L in rainy season. While in village region, it was no difference that nitrate in different season were at the same level. Figure 3 shows the nitrate time distribution.

*: c means in cultivation region; **: v means in village region

Figure 3 nitrate time distribution

3.1.3 Nitrate Change

According to geological form and the aquifer, 11 representative regions were chosen to analyze the nitrate change. Figure 4 shows that nitrate content in upriver position was less than that in backward position and it is obviously that nitrate came from crop field. Since BOD content in groundwater was so low that denitrification couldn’t occur, nitrate accumulated on the way until up above 10mg/L, which is much higher than 1.24 mg/L in Dianchi Lake.

Groundwater level along Dianchi Lake was 1886.84 m, 1885.57 m, 1886.16 m and 1887.22 m, all of which were all above Dianchi Lake water level (1885.50 m), as shown in Figure 4. In this case, groundwater polluted by fertilizer flew into Dianchi Lake by hydrologic grads. Using GMS 4.0 and calculated by water flux and nitrate content, nitrate flew into the Lake was 21.3 t/ (a·hm²), which was 39% of the nitrogen which caused eutrophication of Dianchi Lake (54.6 t/ (a·hm²)).

*: Z1U means Z1 Upper Position; Z1D means Z1 Dianchi Lake

**: Z1U - Z4D are shown in Figure 1

Figure 4 chemical characters of ground water

3.2 Nitrogen balance and transport

In this research, nitrogen was divided into two parts, one of which was Nitrogen Input, including mainly fertilization, and the other of which was Nitrogen Output, including soil withhold, plant use, volatilization (NH₃, N₂O, NO/NO₂ and N₂ etc) and eluviation.

3.2.1 Nitrogen volatilization

3.2.1.1 NH₃

NH₃ volatilization is direct loss of nitrogen and it is affected by soil character, weather and moisture, etc. Because the soils in plot field were acid and neuter soil and conservatory possessed closed what eliminated the effect of wind, main factors that influenced NH3 volatilization comprised cation exchange amount, temperature and moisture level. The results showed that day-temperature diversification influenced visibly the NH₃volatilization efficiency and there were positive correlation between them. NH₃ volatilization velocity was highest in autumn period because there were less nitrate adsorption and more redundant efficient nitrate in growth period than those in autumn period.

Investigation in different season and treatment showed that NH₃ volatilization of Plot I is 0.09~0.72 kg NH₃-N/hm², which were 0.08% (LF) and 0.06% (LF) of total nitrogen input; NH₃ volatilization of Plot II is 2.65~23.44 kg NH₃-N/hm², which were 2.57% (LF) and 1.96% (LF) of total nitrogen input. The main factors that affect NH3 volatilization are osmosis, temperature and moisture content with diversity analysis.

There were obvious correlation in different treatments between NH₃ volatilization and nitrogen fertilization, what behaved volatilization amount in NF treatment was higher than control. This result can be explained that transpiration of soil and crops would accelerate water transpiration, which positively correlated with NH₃ volatilization. There were specific linear correlation between NH3 volatilization and nitrogen fertilization in three treatments, including NF, LF and HF.

3.2.1.2 NOx

NOx discharge ability of soil was related to efficient nitrate content, and discharge amount had relationship with soil characters^[4], climate factors and agriculture activities. NO/NO₂ discharge velocity was laid on soil temperature, which would cause higher discharge velocity when it was higher. In growth period, NO/NO₂ discharge velocity was higher. Dry-wet principle caused by irrigation changed the environment of soil, which enhanced nitrification and denitrification process^[5]. Efficient nitrate fertilizer (carbamide, ammonium bicarbonate) was the main factor of higher NO/NO₂ discharge velocity in growth period.

NOx volatilization loss in different treatments were shown as follows: in Plot I, NOx volatilization loss was 15.45-134.59 kgN/hm2 and the NOx volatilization was 13.86% (LF) and 11.22% (HF) of total nitrogen input respectively while 4.06% and 3.43% in CK and NF treatment; in Plot II, NOx volatilization loss was 18.71-183.97 kgN/hm2, and the NOx volatilization was 19.02% (LF) and 15.33% (HF) while 4.16% and 6.06% in CK and NF treatment. 4.16%和6.06%. Compared with different treatments, the higher nitrogen input, the higher NO/NO2 volatilization.

3.2.1.3 N₂

N₂O produced by denitrification in soil can be deoxidized to N2 by N2O Oxidation Enzyme (N₂OR). Denitrification is in low N2O/N2 condition when field moisture content is 77% ^[7]. Based on locale inspection, field moisture content in research area was between 15% and 29% which is near to the optimal denitrification moisture condition propitious to denitrification^[8]. N₂ volatilization were 31.2~128.4 kg N·hm-2 (LF) and 67.3~276.0 kg N·hm-2 (HF) calculated by NOx output inspection and the value of N₂/NOx under such moisture content.

3.2.2 Plant use

Absorption of efficient nitrate by plants can reduce content of anion nitrate and depress the transform of it, including volatilization, nitrification and denitrification. Since TN content soil in intensive cultivation region is high, nitrate used by plants is only a small part of TN and the effect that the usage of nitrogen in intensive cultivation region in less than in tradition forms. Besides that, the secretions and casts of crop root could enhance microbial amount and bioactivity to improve nitrogen volatilization loss.

The biomass and the nitrogen content of plant with different treatments have no distinct diversity. In Plot I, plant biomass is 7.3% (LF) and 10.7% (HF) higher than NF treatment, while plant nitrogen content is 10.2% (LF) and 11.4%(HF) higher; in Plot II, plant biomass is 8.7% (LF) and 11.4% (HF) higher than NF treatment, while plant nitrogen content is 5.2% (LF) and 14.5% (HF) higher. The results showed that the plant biomass and nitrogen content had no distinct relationship with the fertilization when nitrogen was superfluous in intensive cultivation area. Apparent fertilizer nitrogen usage in LF treatment was 20.8%-30.4%, higher than HF treatment which was 8.5%~11.5%.

3.2.3 Eluviation

The nitrogen input into the system is eluviated into groundwater except soil withhold, plant use and volatilization. The zero flux-plane method was used to calculate vertical infiltration recharge and evaporation-transpiration, which was based on the soil-water hydraulic theory that the water flux did not exist when a gradient of total potential of soil-water was zero. Irrigation intensity was averagely 8.276 mm/d during celery cultivation. The calculated output from the aeration zone was generally 6.783 mm/d, including vertical infiltration recharge and evaporation-transpiration. In other words, 4.38 percent of irrigation was infiltrated into groundwater, and 77.58 percent of irrigation water was evaporated and transpired through plants and surface soils. The derivation between the calculated value and the observed value is 18.04%, which probably resulted from horizontal loss of irrigation water (limited water recharging the ditch was observed during in-situ experiments) and observed precision. In Plot I, nitrogen eluviation is 95.3 kg/ha (LF) and137.2 kg/ha (HF), while in Plot II, it is 174.5 kg/ha (LF) and 239.9 kg/ha (HF).

3.2.4 WSPAC nitrogen balance and transport

The nitrogen balance among water-soil-plant-air continuous (WSPAC) included volatilization, plant use, eluviation and soil withhold, which determined the nitrogen transport and form. The results shows that plants, air, groundwater and the soil are the main components of WAPAC and the nitrogen input can resist or output through these four sections by transport. In WAPAC, nitrogen balance and transport in different conditions are showed in table 2.

Table 2 Losses of the nitrogen in the multisystem

3.3 Factors analysis

Only 30%-40% of nitrogen input (fertilizer) can be absorbed by plant and most of the nitrogen is lost into the environment through different ways, what reduces the nitrogen usage, increases the agriculture cost and causes the pollution of water environment, especially the groundwater environment. Some factors effects a series of physical, chemical and biochemical actions that influence the balance and transport between different forms of nitrogen as well as nitrogen and mediums in WAPAC.

3.3.1 Nitrification Effect

Like ammoniac, nitrate produced in denitrification can be absorbed by plants easily but is prone to eluviating in to the groundwater. Except for the physical and chemical characters of soil and fertilizer (organic or inorganic), the nitrification velocity can also be effected by depth of unsaturated zones and the form of irrigation. Nitrification is related with the depth of soil and the nitrification velocity is highest in surface layer soil, which is 2.50-8.53 mg/(kg·h). Nitrification effect weakens with the depth of soil increases until it reaches 0.6-1.0m, there the nitrification velocity is steady at 0.2-0.3 mg/(kg·h). The relationship of nitrification velocity in surface layer soil is: NF>LF>HF>CK, as shown in Figure 5.

3.3.2 Denitrification Effect

The result shows that the denitrification is related with the depth of soil obviously and the relationship presents saddle shape. Denitrification velocity reduces rapidly from 14.63~31.09 mg/(kg·h) to 5.46~8.72 mg/(kg·h) with the depth of soil increasing from 0.0 to 1.0 m; while denitrification velocity increases from 5.46~8.72 mg/(kg·h) to 6.22~18.94 mg/(kg·h) with the depth of soil increasing from 1.0 to 1.5 m, respectively. That showed denitrification was obviously affected by both nutrient level and DO, and denitrification velocity reduced with nutrient’s decrease above 1.0 m while it increased with DO’s decrease below 1.0 m. At the same depth, denitrification velocity in different is as follows: CK>NF>LF=HF; the number if denitrification bacteria is as follows: CK>NF>LF>HF. Denitrification is restrained by fertilization: the number and the activity of denitrification bacteria is less and the denitrification effect is lower with larger fertilization, as shown in Figure 6.

Fig. 5 Nitrification RateVelocity

 Fig. 6 Denitrification RateVelocity

3.3.3 Ammonia Adsorption Effect

Ammonia is withheld by soil except plant use and nitrification and its adsorption can be calculated by ammonia equation according to the adsorption isothermal figure. The results showed that the ammonia equation accorded with the Freundlich linear adsorption isothermal mode and the adsorption coefficient reduces with the soil depth increasing. Ammonia adsorption affect mainly occurred in surface layer soil and the Adsorption quantity reduces when the soil depth increased.

4 Summary

In this study, local investigations and in-situ experiments were carried out in this research to investigate the nitrogen balance among water-soil-plant-air continuous (WSPAC) and the control factors of nitrogen balance and transport through unsaturated zones of soils. It was found that a number of environmental factors influenced the balance of nitrogen in groundwater. The experimental results can be summarized as follows:

High contaminant of nitrate in groundwater was related to the degree and years of cultivation and 39% of the nitrogen which caused eutrophication of Dianchi Lake came from groundwater stream.

The quantity of NO₃-N and NO₂-N recharged into groundwater were related to fertilization. The values of vertical fluxed into groundwater of NO₃-N were 19.30 (HF) and 5.52 (LF) g/plot and those of NO₂-N were 0.96 (HF) and 0.73 (LF) g/plot.

The vertical infiltration recharge was limited by groundwater level and hydraulic conductivity of soil and this can be explained by higher withhold of soil in thicker unsaturated zone.

NO SIGNIFICANT DIFFERENCE OF PLANT USAGE WAS FOUND AMONG TREATMENTS OF 93.5-138.1 KG N/HM², WHILE THE PERCENTAGE OF NITROGEN INPUT WAS QUITE DIFFERENT, 20.8-30.4% IN LF, HIGHER THAN 8.5-11.5% IN HF.

Nitrogen volatilized into the air is related to the denitrification ability stimulated by organic matter supply and the ammonia volatilization relevant to cation exchange and evaporation in soil. Redox potential has also been shown to affect denitrification rates, with higher rates in more reduced treatments. Denitrification rate in unvegetated treatment was 6.36 mg/(kg·h), higher than 2.79 mg/(kg·h) in vegetated one.

Vertical infiltration recharge and evaporation-transpiration rate were separately 0.363 mm/d in thick unsaturated zone while 1.851 mm/d on the other condition. The quantity of NO₃-N and NO₂-N recharging into groundwater are related to unsaturated zone. The values of vertical fluxing into groundwater of NO₃-N were 19.30 g/plot and those of NO₂-N were 0.96 g/plot in thicker unsaturated zone, while 108.52 g/plot of NO₃-N and 0.11 g/plot NO₂-N in thinner one, respectively.

Acknowledgements

I would like to acknowledge the field and laboratory work of Environment Institute, Kunming City. This research is supported by National Nature Fund (40372107) and Tackle Key Problems of the Tenth Five-year Plan.

References

[1]         Hancock CM, Rose JB, Vasconcelos GJ, Harris SI, Klonicki PT, Sturbaum GD. Giardia and Cryptosporidium occurrence in groundwater. J Am Water Works Assoc 2000, 92 (9):117-23.

[2]         Z. Feleke, K. Araki, Y. Sakakibara, T. Watanabe, M. Kuroda, Selective reduction of nitrate to nitrogen gas in a biofilm-electrode reactor, Wat. Res. 1998, 32 (9):2728-2734.

[3]         Jin ZF, Chen YX, Wang F, Ogura N, Detection of nitrate sources in urban groundwater by isotopic and chemical indicators, Hangzhou City, China, Env. Geo. 2004, 45 (7): 1017-1024.

[4]         Musters, P. A. D. Bouten, W. A method for identifying optimum strategies of measuring soil water contents for calibrating a root water uptake model. Journal of Hydrology, 2000, 227: 273-286.

[5]         Brandelik, A. & Hubner, C. Soil moisture determination-accurate, large and deep. Phys. Chem. Earth. 1996, 21(3): 157-160.

[6]         Yan Jin, Thilo Streck, William A. Jury Transport and biodegradation of roluence in unsaturated soil. Journal of Contaminant Hydrology, 1994, 111-127.

[7]         Ridley AM, Mele PM, Beverly CR, Legume-based farming in southern Australia: developing sustainable systems to meet environmental challenges, Soil Biology & Biochemistry, 2004, 36 (8): 1213-1221.

[8]         Seunghee Choe, Yoon-Young Chang, Fyung-Yub Hwang, JeEhyeong Khim, Kinetics of reductive denitrification by nanoscale zero-valent iron, Chemosphere 2000, 41: 1307-1311.

[9]         Davidson, E. A. Fluxes of N₂O and NO from terrestrial ecosystems[A]. Rogers J E and Whitman W B, ed. Microbiol Production and Consumption of Greenhouse Gases: CH4, NOX and halomethanes[M].Washington DC: Am. Soc. Microbiol., 1991, 219-235.

[10]     Hansen S, Mæhlum J E, Bakken L R. N₂O and CH₄ fluxes in soil influenced by fertilization and tractor traffic [J]. Soil Biol Biochem, 1993, 25:621-630.