BACKGROUND With the increase of population, urbanization development, the large discharge of industrial wastewater and the excessive application of agricultural nitrogen fertilizer, the problem of “three nitrogen” (nitrate nitrogen, ammonia nitrogen, and nitrite nitrogen) pollution in groundwater has become increasingly serious. Due to the slow flow, weak alternation degree and poor self-purification ability, the nitrogen polluted groundwater is difficult to rehabilitate and the repair cost is high. It is of great practical significance to carry out prevention and control from the pollution source. With the concealment, complexity and hysteretic groundwater pollution property, it is difficult to conduct the source analysis of nitrate in groundwater. Positive matrix factorization (PMF) model, as a new type of source analysis model, has more interpretable and clear physical significance in terms of factor load and source factor score. At present, the application of the PMF model in the source analysis of “three nitrogen” in groundwater media is rarely reported, and the research in this field needs to be deepened. Gas-phase molecular absorption spectrometry is a practical analytical method for nitrogen compounds with high accuracy, wide linear range without color and turbidity interference. This method does not require chemical separation, and uses few chemical reagents and innoxious reagents. At present, this method has been widely used in the analysis and determination of water quality, nitrite nitrogen, ammonia nitrogen, nitrate nitrogen, total nitrogen, and other substances.

OBJECTIVES To identify the source and distribution characteristics of the “three nitrogen” in shallow groundwater in the Hailun area.

METHODS (1) Determination of “three nitrogen” and other elements in groundwater samples. With GPS coordinates, the 40 shallow groundwater samples were collected in Helen River Basin and Zaying River Basin. The collection and preservation of groundwater samples were performed in accordance with the environmental standards HJ/T 164—2004, HJ/T 195—2005, HJ/T 197—2005 and the ecological industry standards DZ/T 0064—2021. The quality control of the “three nitrogen” test in groundwater was based on the requirements of the national standard GB/T 14848—2017, the technical requirements for analytical quality control of groundwater pollution investigation (DD 2014-15), and the specification of testing quality management for geological laboratories (DZ/T 0130—2006). Each batch contained twenty samples and the test process of each batch was equipped with laboratory blank, reference materials, laboratory duplicate samples and quality control of external monitoring samples. All of the laboratory blank results were less than 2 times the detection limit of the method. The added standard recovery rate of the sample matrix was between 80%-120%, with the relative deviation (RD) of the laboratory repeat samples under 15%. The qualified rate of external quality control sample was 100%.　　(2) Analysis of detection data. Statistics and analysis of “three nitrogen”, heavy metal content correlation and principal components in groundwater were carried out using WPS Office 2016 and SPSS 22.0. The spatial distribution map of “three nitrogen” content in groundwater was obtained by CoreDRAW X5.0 software. The Ⅲ class water criterion in Standard for Groundwater Quality (GB/T 14848—2017) was used as the criterion for drinking water source and industrial and agricultural water. The Nemerow pollution index was calculated and compared with the corresponding grade standard index. The PMF model was used to identify and quantify the contribution rate of each nitrogen source to comprehensively judge the source of “three nitrogen”.

RESULTS (1) The contents of “three nitrogen” and other elements in the groundwater samples in the study area were determined. According to the contents of “three nitrogen” and related parameters shown in Table 3, the detection rates of the nitrate nitrogen, ammonia nitrogen and nitrite nitrogen were 100%, 77.5% and 90.0%, respectively. The maximum detection content of nitrate nitrogen was 123.05mg/L, and average content was 15.27mg/L. The maximum detection content of ammonia nitrogen was 3.91mg/L, and average content was 0.33mg/L. The maximum detection content of nitrite nitrogen was 0.65mg/L, and average content was 0.12mg/L. Compared with the Ⅲ class water criterion, the nitrate nitrogen over standard rate was 20%, with the maximum value 6.18 times; the ammonia nitrogen over standard rate was 12.5%, with its maximum value 7.8 times; the nitrite nitrogen did not exceed the Ⅲ class water criterion. The content of Mn was ND-3.20mg/L, and the proportion of Mn in I-III class water range was 40%. The contents of 8 heavy metals including Cd were within 10 times the detection limit, with 100% Ⅰ-Ⅲ class water. The content of anion F⁻ was 0.029-0.69mg/L. The content of anion Cl⁻ ranged from 0.0027 to 310mg/L, with the 97.5% Ⅰ-Ⅲ class and 2.5% over standard rate. The content of SO₄ ²⁻ was 0.46-433mg/L, with the 95% Ⅰ-Ⅲ class and 5% over standard rate.　　(2) The pollution and spatial distribution of groundwater were studied. The results of Nemerow comprehensive pollution evaluation showed that the pollution degree of “three nitrogen” in the study area was relatively low. The proportion of unpolluted samples, mild pollution samples, heavy pollution samples and serious pollution samples was 65%, 25%, 2.5% and 0, respectively (Table 4). “Three nitrogen” spatial distribution is shown in Fig.1a-c. Samples with ammonia nitrogen content ranging in Ⅰ-Ⅲ class water was mainly concentrated in the north, west, east and central; the Ⅳ class water samples were concentrated in the middle; the class V water samples were concentrated in the center. Fig.1a shows the trend of high ammonia nitrogen in the middle and low ammonia nitrogen on both sides. In Fig.1b, Ⅰ-Ⅲ class water samples of nitrate nitrogen were mainly concentrated in the central, east and north, with the Ⅳ class water in the middle and the class V water in the west. The samples with the highest and sub highest nitrate nitrogen content were located in the center of the study area, radially distributing along the Helen River Basin and the Zaying River Basin. The nitrate nitrogen content was high near the central area, with the low content far from the central area. The nitrite nitrogen in Fig.1c was Ⅰ-Ⅲ class water. The spatial distribution of the nitrite nitrogen shows the trend of high in the north and low in the south along the Helen River Basin.　　(3) Multivariate statistical analysis and source of “three nitrogen” in groundwater were studied. Multivariate statistical analysis of Pearson correlation coefficient showed that nitrate nitrogen had some homology with Cd, Co, Ni. Nitrate nitrogen in groundwater was significantly correlated with Cl⁻ and SO₄ ²⁻, while ammonia nitrogen and nitrite nitrogen showed no correlation with anion. Sample detection data imported into EPA PMF5.0 software for PMF source analysis, and the industrial source, agricultural source, living source and natural factor source were taken as the four source analysis factors of “three nitrogen” in the groundwater in the study area. The PMF source analysis results are presented in Table 6 and Fig.2. The main sources of “three nitrogen” in the groundwater in the study area were the compound source of living and industrial source, which further indicated that human activity was the root cause of “three nitrogen” pollution in the groundwater in Helen area. According to the spatial distribution of “three nitrogen” in groundwater, the highest point of nitrate nitrogen pollution was distributed in the center of the research area, with the relatively developed human life and industrial activities in this area. The ammonia nitrogen pollution samples were related to the dense human activities in the middle of the study area. The over-standard area of nitrate nitrogen was higher than the ammonia nitrogen, which indicated that the compound source of living and industrial source was the main factor affecting the “three nitrogen” content in the groundwater in study area.

CONCLUSIONS The nitrate nitrogen and ammonia nitrogen are the nitrogen pollution components in the groundwater in the study area. Compared with the water index in groundwater quality standard (GB/T 14848—2017), the over standard rates of the nitrate nitrogen and ammonia nitrogen are 20% and 12.5% respectively. The groundwater quality with pollution evaluation grade I （unpolluted） to grade Ⅲ （moderate polluted）accounts for 92.5%, which indicates relatively light pollution.　　The spatial distribution of ammonia nitrogen in the study area shows a trend of high in the middle and low on both sides. Nitrate nitrogen and ammonia nitrogen pollution samples are mainly distributed in the center of the intensive human activity research area. The quantitative analysis results of the “three nitrogen” PMF analysis model show that the compound source of living and industry are the main source of nitrate nitrogen pollution in the study area, and the compound source of living and agricultural production is the main source of ammonia nitrogen pollution. Combining the situation with the source of the “three nitrogen” pollution in the research area, the project of the “three nitrogen” pollution prevention can be carried out to strengthen protection, sustainable development and utilization of groundwater resources in the Helen area.