Introduction

The rapid population growth, urbanization, industrialization, and the rise of living standards have contributed to environmental pollution, especially in drinking water supplies (Zarei and Bahrami 2016; Mokarram et al. 2015b, c; Naseri et al. 2006). The development process in Iran has caused many problems including water pollution; this issue becomes more important since Iran is placed in an arid and semi-arid area and groundwater supplies about 52% of drinking water (Mokarram et al. 2015c; Beiranvand et al. 2014). Regarding the lack of water and drought crises in recent years, the importance of groundwater has been increasing day by day mainly in the western United States (Turner et al. 2019) and Asian countries such as Iran (Nosrati and Zareiee 2011; Bahrami et al. 2017, 2018; Abou Zaki et al. 2019), India (Voss et al. 2013; Mukherjee et al. 2018), Bangladesh (Mojid et al. 2019), China (Werner et al. 2013), and Saudi Arabia (Michel et al. 2012).

The greatest threat to the groundwater in the future is the pollution of this source by harmful substances entered by humans deliberately or unintentionally or as a result of negligence and ignorance. Nitrate is the most abundant and perhaps the most common pollutant that threatens groundwater. Nitrate is the most prevalent groundwater contaminant detected specifically in shallow aquifers because of both point and nonpoint sources (Esmaeili et al. 2014). This ion in drinking water has two undesirable health effects including methemoglobinemia or “blue baby” syndrome in infants and the potential of carcinogenic compounds in adults (Schipper and Vojvodić-Vuković 2000). Liu et al. (2017) assessed the concentration and spatiotemporal distribution of groundwater nitrate under cropland by statistical and geostatistical techniques in Shandong province, China. Results revealed that the median nitrate concentrations after the rainy season were considerably more than those before the rainy season, and reduced with enhancing groundwater depth. Also, nitrate under the vegetable and orchard area is considerably more than the ones under grain. The kriging map indicated that groundwater nitrate has a powerful spatial variation.

During the past two centuries, production and consumption of nitrate, especially in the agricultural field, have been increased considerably. Currently, many countries in the world including Iran (Badee Nezhad et al. 2017), India (Devendra et al. 2016), China (Meng et al. 2018), etc., encounter a high nitrate level in drinking water. The main reason is the recharge of agricultural runoff and municipal and industrial wastewater to water resources, particularly groundwater. Ducci (2018) presented a technique for evaluating nitrate pollution susceptibility in groundwater applying thematic maps. The results of this paper showed a partly good correlation coefficient between the created nitrate pollution susceptibility map and the nitrate distribution in groundwater in the study area. Han et al. (2018) assessed the sources and processing of nonpoint sources of nitrate in China. The results of the spatial analysis indicated that NO3 concentration was great in the tea plantation and forest zones, and in the temporal analysis showed a high NO3 level in spring.

Green et al. (2018) assessed the zonal variation of nitrate fluxes in the unsaturated zone and groundwater in Wisconsin. Results indicated that under common activities and conditions, almost 60% of the shallow aquifer will finally be impressed by the downward migration of NO3−, with denitrification protecting the remaining 40%. Akpan et al. (2018) evaluated the spatial distribution of pollutants and their values in soil and water resources applying geophysical and geological data include pH, EC, TDS, Cl−, NO3−, HCO3−, HCO3−, SO42−, Ca2+, Na+, K+ and Mg2+ in Nigeria. The findings indicated that ionic concentrations in the sand-dominated soils and water were within permissible limits and baseline standards.

In recent years, numerous studies have been conducted to investigate the nitrate pollution in the water resources (Chand et al. 2011; Maghanga et al. 2013; Ameur et al. 2016; Hoseinzadeh et al. 2016; Akale et al. 2018; Sun and Chai 2018; Jahangeer Gupta and Yadav 2018; Ducci et al. 2019; Mir et al. 2019; Ogrinc et al. 2019; Rawat et al. 2019; Yin et al. 2019). Based on the conclusion of researches done in Iran, nitrate contamination in the water resources of Iran is at a medium level. In most of the researches, high nitrate concentrations are because of lack of sewage collection networks, discharge of urban and industrial sewage to water resources, and agricultural practices that apply high values of manure and fertilizer (Akhavan et al. 2014). Therefore, the present study is aimed to assess the temporal and spatial trend of nitrate contamination in Fasarud Plain groundwater, southern Iran, using GIS and demonstrate the achieved findings by using nitrate pollution index (NPI).

Materials and methods

This study utilized groundwater quality data collected from 31 water wells of Fasarud Plain, Darab County, Iran, from April 2017 to March 2018 (in total, 279 samples). In each well, the samples were taken after a pumping period of 10 min at least. The water samples were collected in polyethylene bottles. They were stored in the refrigerator until they were analyzed (Mousavi and Amiri 2012). The nitrate concentration was measured by spectrophotometry method (APH 1998; Wang et al. 1998). For this, an aliquot (50 mL) of water sample was transferred to a 100 ml beaker, 1 mL HCl (1.0 N) was added. The absorbance of the sample was measured at 220 nm (NO3 + organic matter) by spectrophotometer. Then, the absorbance of the sample was measured at 275 nm (organic matter). If Abs275 nm × 2 < 10% Abs220 nm, the Abs275 nm × 2 was subtracted from Abs220 nm to obtain value for NO3 corrected for organic matter content.

Validation of chemical analysis results was performed by the ionic balance verification and by repeating the analysis for the same sample. The precision of the results was estimated by calculating the charge balance error (CBE). The test results are considered reliable only when the CBE is less than or equal to 5% (Guangwei, 2013). The limit of nitrate content in drinking water has set by the World Health Organization (WHO 2011) and also the Institute of Standards and Industrial Research of Iran (ISIRI, 2009) at 50 mg L−1. Based on the achieved findings for the nitrate concentration, water quality was classified into three classes as Table 1.

Table 1 Values and categories of NPI and WHO limits on NO3 (WHO 2011; Almasi et al. 2016; Obeidat et al. 2012)
Full size table

Then, nitrate contamination of the studied area was assessed by the nitrate pollution index (NPI). To evaluate the nitrate changes over the district within a year, spatial and temporal trends analysis were employed.

Study area

The Fasarud Plain is placed in the Fars Province (the south part of Iran) and extends geographically between 54° 13′ and 54° 32′ East longitude and from 28° 39′ to 28° 49′ North latitude (Fig. 1). The total area of the studied region is 7500 km2 with an altitude of 1180 m. According to the De Martonne index, the climate of Fasarud Plain is semi-arid, the mean annual temperature is about 25 degrees Celsius, the mean annual rainfall is 350 mm, the mean annual wind speed is about 1.2 m s−1, the mean annual potential evapotranspiration is about 1821.1 mm, and the mean annual sunshine is about 9.4 h day−1.

Fig. 1
figure 1

Land use map and groundwater flow direction of studied area with spatial distribution of the studied wells

Full size image

According to Fig. 1, the most area of the Fasarud Plain is composed of alluvial formation that resulted in the agricultural occupations using groundwater resources as irrigation water supply. The Korsia salt diapir (Sp) is specified in the north of plain. Throughout the north plain including around the salt diapir is formed almost exclusively by Tarbur limestone (Tb), which is a predominantly carbonate lithostratigraphic unit that outcrops in Zagros basin (Maghfouri-Moghadam et al. 2009). The Asmari limestone (As) formation is represented in east and south of plain, that is of Late Oligocene (Chattian) Early Miocene (Burdigalian) age and the youngest source rock in Zagros (Maghfouri-Moghadam and Khanjani 2014). The Upper Cretaceous–Paleocene Gurpi Formation (Cpm) consists of deep-marine shales, marls, and argillaceous lime mudstones that crop out in the Zagros Mountains (Beiranvand et al. 2014), is represented in west and northwest of Fasarud Plain.

Nitrate pollution index

A single-parameter water quality index named the nitrate pollution index (NPI) was applied for measurement of the nitrate pollution in the investigated wells. Applying the NPI index is indicative of nitrate pollution in the groundwater due to anthropogenic activity. The following formula was applied to compute the NPI (Obeidat et al. 2012):

$${\text{NPI}} = \frac{{C_{{\text{s}}} - {\text{HAV}}}}{{{\text{HAV}}}}$$
(1)

where Cs is the nitrate concentration in the sample, and HAV is the threshold value of the anthropogenic source (human-affected value), taken as 20 mg L−1 (Obeidat et al. 2012). According to the achieved findings for the NPI, water quality was classified into five categories as Table 1.

Temporal trend analysis

Plenty of statistical methods (parametric or nonparametric) have been developed to find trends within time series such as linear regression, Spearman’s Rho test, Mann–Kendall test, Sen’s slope estimator, and Bayesian method. In the present research, after assessing the normality of nitrate data series by Shapiro-Wilk method (the variables of normality tests were the stations), the trend of these series was evaluated based on the Pearson method (parametric method) and the Spearman’s Rho test (nonparametric method), for normal and non-normal values of nitrate, respectively.

Pearson test

Pearson r correlation is the most extensively applied correlation statistic to estimate the degree of the relationship between linearly related parameters. Pearson r correlation is applied to calculate the degree of relationship between the two parameters. The following formula is applied to measure the Pearson r correlation:

$$r = \frac{{N\sum{xy} - \left( {\sum x } \right)\left( {\sum y } \right)}}{{\sqrt {\left[ {N\sum{x^{2} } - \left( {\sum x } \right)^{2} } \right]\left[ {N\sum {y^{2} } - \left( {\sum y } \right)^{2} } \right]} }}$$
(2)

where N is the number of observations, X is the independent variable, Y is dependent variable, and r is correlation between X and Y.

Spearman’s Rho test

Spearman’s Rho test is a nonparametric technique customarily applied to verify the absence of tendencies. This test is a statistical measure of the strength of a monotonic relationship between paired data. At Spearman’s Rho test, the positive quantities of correlation coefficient show the upward trend in data series and the negative quantities of correlation coefficient show the downward trend in data series. In this test, statistic \({D}\) is represented as follows (Sneyers 1990):

$$D=1-\frac{6\sum_{i=1}^{n}{(R\left({X}_{i}\right)-i)}^{2}}{n({n}^{2}-1)}$$
(3)

where \(R\left({X}_{i}\right)\) is the rank of the observation, \({X}_{i}\) is in the time series and, n is the length of the time series.

Spatial trend analysis

There are many interpolation methods for the analysis of the spatial distribution of different parameters, such as geostatistical analysis including inverse distance weighting (IDW), Kriging, and Co-Kriging with Circular, Gaussian, Spherical, and many other variograms (Nazaripur et al. 2015). In the current study, after assessing the nitrate data normality using the Shapiro-Wilk method (using SPSS Software), ArcGIS10.2 interpolation techniques were used to prepare nitrate spatial distribution maps in the studied area (from April 2017 to March 2018 on a monthly basis). Hence, inverse distance weighting (IDW) geostatistical wizard (with pixel size equal to 10 × 10 m) was used to interpolate the data series of nitrate that was non-normal (\({\text{sig}}{.} < 0.05\)), but for those series of data that was normal, various geostatistical wizards were tested to find the appropriate interpolation method. In this regard, the error criterion of root means square error (RMSE) was used to assess the accuracy and precision of various interpolation methods. Finally, area percentages of each three nitrate classes were extracted from the prepared spatial maps.

After the preparation of nitrate maps and calculation of area percentages for each nitrate class, the normality of data series of area percentages over time was assessed using Shapiro-Wilk’s statistical test. Then, the trend of the change of area percentages of different nitrate classes over time (monthly scale) was assessed based on statistical parametric (Pearson) and nonparametric (Spearman’s Rho) tests, for normal and non-normal values of data, respectively.

Results and discussion

The statistics of NO3 and NPI parameters based on all wells are represented in Table 2. The minimum value of nitrate concentration was observed at Well 22 located at the east of plain in June (2.43 mg L−1) and maximum value allocated to Well 16 located in the same area in March (96 mg L−1). The minimum value of NPI was observed at Well 22 in June (− 0.88) while the maximum value allocated to Well 12 in the northwest in September (3.80). Based on the agricultural use of the lands around these wells, the maximum values of nitrate concentration and NPI were observed in months that irrigation was practiced and the minimum quantities were in the months that irrigation had been stopped. Also, according to the mean value of NO3 concentration equal 41.32 mg L−1 and the mean value of NPI equal 1.07, it can be concluded that the pollution of nitrate at Fasarud Plain is in the desirable and moderate range.

Table 2 Descriptive analysis of nitrate (mg L−1) and NPI in the studied wells
Full size table

Temporal trend analysis of nitrate

Values of normality using the Shapiro-Wilk test and consequent temporal trend using the Pearson method and the Spearman’s Rho test for normal and non-normal values of nitrate, respectively, from April 2017 to March 2018 represented in Table 3.

Table 3 Normality and temporal trend tests of nitrate (from April 2017 to March 2018)
Full size table

Nitrate concentration in most wells followed the normal distribution except the Wells 18, 19, 22, and 26. Besides, the temporal trend of nitrate concentration in most wells was positive that means the nitrate concentration in most wells is ascending from April 2017 to March 2018. The highest enhancement of nitrate took place in Wells 1, 6, 9, 11, 13, 15, 16, 18, 26, and 28 at the significance level of 1%, and in the next rank in Wells 4, 20, 24, and 31 at the significance level of 5%. Besides, in Wells 8, 19, 21, and 29, the nitrate concentration trend was descending that only Well 29 had a significant decrease at the significance level of 1%.

The Wells had the increasing temporal trend of nitrate, are on the way of groundwater dominant flow or have the formations of Gurpi, Asmari, or Tarbur. Besides, well 29 with the significant decreasing trend of nitrate at 1% level is in the vicinity of salt diapir of Korsia. These results show that the reason for nitrate changes from April 2017 to March 2018 cannot be geological formations. Therefore, the agricultural drainage can be the main reason for nitrate increasing trend in the studied wells. Liu et al. (2017) resulted that the temporal variability of groundwater nitrate in Shandong Province, China, had a certain trend over the years. Mohammadi et al. (2017) based on the zoning maps of groundwater in Bandar-e Gaz City, Iran, found that in the dry seasons, nitrate and hardness concentration is more than rainy seasonal.

Spatial trend analysis of nitrate

Nitrate concentrations ranged between 2.43 and 96 mg L−1. The most values that characterize the waters of the western, eastern, southern, and northwestern areas of the aquifer relate to the groundwater flow direction and fertilization of agricultural lands. The spatial distribution of NO3 concentration is represented in Fig. 2. Nitrate is a very essential factor for evaluating the contamination of groundwater resources. This parameter is related to the desirable limit of 50 mg L−1 in 194 samples (70%). No water sample is within the maximum permissible limit. In the survey region, 85 water samples (30%) exceed the nitrate concentration limit (50 mg L−1).

Fig. 2
figure 2

Spatial distribution map of nitrate (from April 2017 to March 2018)

Full size image

The results of normality test using the Shapiro-Wilk method and consequent spatial trend using the Pearson method and the Spearman’s Rho test for normal and non-normal values of area percentage of nitrate classes, respectively, from April 2017 to March 2018 represented in Table 4. According to results, the area percentage of classes 1 (desirable limit) and 3 (not permissible limit) followed the normal distribution at the significance level of 5% and class 2 (maximum permissible limit) did not have any records throughout the plain.

Table 4 Normality and spatial trend tests of area percentage of nitrate classes
Full size table

Also, the spatial trend of area percentage of nitrate classes was negative and positive for classes 1 and 3, respectively. These results mean that from April 2017 to March 2018, the area percentage of class 1 (in desirable range) is descending and conversely the area percentage of class 3 (in not permissible range) is increasing. In general, the spatial trend of nitrate concentration in Fasarud Plain groundwater deteriorated from April 2017 to March 2018. Liu et al. (2017) showed that the spatial semivariogram can be impressed by intrinsic (physical, chemical, and biological characteristics of hydraulic and geographic conditions) and/or extrinsic (agricultural management practices, such as fertilization, irrigation, and animal wastes) parameters.

The NPI ranged between 0.88 and 3.8. According to the spatial distribution of NPI shown in Fig. 3, the lowest values of NPI related to clean water in terms of anthropogenic nitrate are observed in areas around the Wells 10, 24, 32, and 33. The light nitrate pollution based on NPI values is dominant over the most area especially from north to south of central areas during the most studied period. The NPI values demonstrated the anthropogenic nitrate pollution at moderate, significant, and very significant levels are most common in the northwest and eastern regions that vary over the irrigation season. That way, these areas are reduced from April 2017 to July 2017 due to diminishing the irrigation practices, but beginning the irrigation period from August or September 2017 causes to increase these classes of NPI.

Fig. 3
figure 3

Spatial distribution map of NPI (from April 2017 to March 2018)

Full size image

According to the NPI and the nitrate concentration, wells were categorized into five classifications (lower than 20 mg L−1, 20–40 mg L−1, 40–60 mg L−1, 60–80 mg L−1, and more than 80 mg L−1). About 24% of the samples had a nitrate concentration that lowered the threshold amount of the anthropogenic source or the human-affected amount (20 mg L−1). About 30% of the sampled sites had a nitrate concentration of greater than 20 mg L−1 and less than 40 mg L−1 (light pollution), while 19% of the sampled sites had a nitrate concentration of greater than 40 mg L−1 and less than 60 mg L−1 (moderate pollution), 16% of sampled sites had a nitrate concentration of greater than 60 mg L−1 and less than 80 mg L−1 (significant pollution), and 11% of samples had a nitrate concentration of greater than 80 mg L−1 (very significant pollution).

Based on similar studies in other parts of the country in recent years, it was recognized that nitrate had an increasing trend, like in the study of temporal and spatial variations of drinking water sources of Gachsaran, Iran, using GIS (Yousefi et al. 2013). Mozafarizadeh and Sajadi (2014) investigated the chemical pollution of Borazjan’s groundwater, Iran, and presented high nitrate pollution up to 160 ppm, especially in the southern Borazjan plains which were occurred by farming activity and absorbing wells.

But Almasi et al. (2016) resulted that there was not a serious problem with nitrate and nitrite concentrations in groundwater of Dehloran and pollutant levels were less than standards. The soils in Dehloran are well-drained, so they have a low capacity to hold water; hence, these soils require some of the highest utilization of fertilizer and irrigation. The results of the current research are in agreement with Alighardashi and Mehrani (2017), who concluded that water use in agricultural areas and the subsequent consumption of fertilizer are the main sources of groundwater contamination with nitrate in Iran. Esmaeili et al. (2014) indicated that nitrate leaching is especially important in regions of intensive farming of Isfahan suburb, but municipal sewage effluents and industrial wastewaters are further sources of nitrate pollution in the adjacent of urban and industrial croplands. They concluded that the nitrate amount of groundwater in the survey area has considerably enhanced because of arid and semi-arid climatic conditions, the indiscriminate exploitation of groundwater, recent droughts, high usage of fertilizers, and enhance of urbanization and industrialization.

Liu et al. (2017) proved that the significant variables for nitrate variance are livestock per unit area, annual mean temperature, vegetable yield per unit area, percentages of irrigation areas, population per unit area, percentages of orchard area, per capita agricultural production, and unit-area nitrogen fertilizer. Generally, statistical analysis indicated that there are many impressing parameters for nitrate variation in space and time (Liu et al. 2017). Deficit irrigation has been suggested as an alternative to decrease nitrate leaching (Amiri et al. 2016). On the other hand, using high nitrate water resources for irrigation can decrease the requirement for inorganic fertilizers and decrease the cultivation cost and nitrate contamination (Esmaeili et al. 2014).

Conclusion

Assessing the temporal and spatial trend of nitrate contamination and nitrate pollution index (NPI) in Fasarud Plain groundwater, southern Iran from April 2017 to March 2018 indicated that increasing temporal trend of nitrate is because of fertilizers using in agricultural lands. As, in most wells, the nitrate concentration and the NPI were increased temporally.

The spatial trend of area percentage of nitrate class 1 (desirable limit of less than 50 mg L−1) was decreasing during the studied period, while its own for class 3 (not permissible limit of more than 50 mg L−1) was increased. In general, results demonstrated that by finishing the irrigation season (about April 2017), nitrate concentration and the NPI reduced temporally in the samples and the percentage area of nitrate class 3 decreased gradually. Beginning the agricultural season (about September 2017), again the NPI, the nitrate concentration, and the percentage area of nitrate class 3 began to increase. Temporal and spatial variations of NPI are corroborated of anthropogenic nitrate pollution in the northwest, eastern, and south of the plain that vary over the irrigation season. As regards groundwater in this plain is also used for drinking purposes, the results of this study can be a warning to water resources authorities and decision makers.