Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/taylor-francis/spatial-distribution-and-potential-ecological-risk-assessment-of-heavy-9b7lwy3EyW
References
References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.
- Publisher
- Taylor & Francis
- Copyright
- © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
- ISSN
- 2474-9508
- DOI
- 10.1080/24749508.2019.1587588
- Publisher site
- See Article on Publisher Site
Abstract
GEOLOGY, ECOLOGY, AND LANDSCAPES 2020, VOL. 4, NO. 2, 87–103 INWASCON https://doi.org/10.1080/24749508.2019.1587588 RESEARCH ARTICLE Spatial distribution and potential ecological risk assessment of heavy metals in agricultural soils of Northeastern Iran a b Ali Keshavarzi and Vinod Kumar a b Laboratory of Remote Sensing and GIS, Department of Soil Science, University of Tehran, Karaj, Iran; Department of Botany, DAV University, Jalandhar, India ABSTRACT ARTICLE HISTORY Received 13 December 2018 Illustrating the spatial distribution and potential sources of soil properties, and heavy metals, Accepted 23 February 2019 viz., Fe, Zn, Mn, and Cu, are the vital prerequisites for decreasing their pollution. The 68 composite agricultural soil samples in triplicates were collected by employing grid method to KEYWORDS evaluate the concentration of pH, sand, silt, clay, organic carbon, P, K, and heavy metals. Agricultural soils; Multivariate techniques (Pearson’s correlation, heatmap, principal component analysis, and multivariate techniques; nonmetric multidimensional scaling), geostatistcal techniques, and contamination indices heatmap; kriging; ecological were employed. The contents of Fe, Zn, Mn, and Cu were lower than the limits for Iran EPA risk assessment guidelines and Earth’s crust. The results of contamination factor and potential ecological risk index (RI) showed that agricultural soils have less contamination and low ecological risks. The enrichment factor (EF), geoaccumulation index (Igeo) and modified ecological risk index (MRI) indicated that 99%, 86.7% and 52.4% agricultural soil samples showed very high enrichment and ecological risks of heavy metals. Both anthropogenic activities and natural factors were responsible for heavy metal contents. The results of geostatistcal analysis revealed that Zn is accumulated more in Central regions, whereas Cu and Mn accumulated more in South and Northeastern regions of the studied area for EF, Igeo, and modified potential ecological RI. Introduction industrial activities were responsible for heavy metal contamination in the soil. Further, the results of Soil is important nonrenewable resources that act as spatial analysis indicated that the distribution of origin and pool of various contaminants. Soil con- heavy metals was affected by human activities and tamination by heavy metals is an important issue, and natural factors. Kelepertzis (2014) while working on various activities such as agricultural, urbanization, heavy metals content in agricultural soils of and industrialization are responsible for enhancing Peloponnese, Greece found that anthropogenic activ- the heavy metal content in the soil (Hu et al., 2013; ities were greatly responsible for heavy metals con- Keshavarzi & Kumar, 2018; Kumar et al., 2019; Zou, tent. Further, the results of geostatistical analysis Dai, Gong, & Ma, 2015). The content of heavy metals showed that high contents of Cu, Mn, and Zn were in the agricultural soils is a matter of great apprehen- attributed to citrus soils cultivated for the production sion due to their accumulative and nondegradable of oranges and mandarins. characteristics (Facchinelli, Sacchi, & Mallen, 2001). The determination of heavy metal concentrations The pollution of heavy metals in agricultural soils was will not provide the degrading effect of heavy metals studied all over the world, including Iran, which is in the environment. In order to determine the pollu- the main issue regarding the possibility of metal tion and ecological risks of heavy metals, various absorption by food (Mohammadi et al., 2018; Tian, indices were applied. Scientists applied various fac- Huang, Xing, & Hu, 2017; Zhang et al. 2018). tors, i.e., contamination factor (CF), enrichment fac- Masoud, El-Horiny, Atwia, Gemail, and Koike tor (EF), geoaccumulation index (Igeo), ecological (2018) while working on soils of Dakhla Oasis, risk index (RI), and modified ecological risk index Egypt applied multivariate and geostatistical techni- (MRI) for assessment of pollution and ecological risks ques and reported that urban and agricultural activ- (Ahmed et al., 2016; Kumar, Sharma, Minakshi, ities were responsible for degrading the soil quality. Bhardwaj, & Thukral, 2018; Tian et al., 2017). The Liu et al. (2017) used multivariate techniques, con- EF and Igeo are based on relative assessment of heavy tamination indices, and geostatistical techniques in metals in polluted and unpolluted soil conditions tobacco growing soils of Shandong, China and (Kumar et al., 2018; Sakram, Machender, Dhakate, inferred that agricultural activities such as application Saxena, & Prasad, 2015). Various researchers have of fertilizers, pesticides, irrigation water, etc., and CONTACT Ali Keshavarzi alikeshavarzi@ut.ac.ir Laboratory of Remote Sensing and GIS, Department of Soil Science, University of Tehran, P.O. Box: 4111, Karaj 31587-77871, Iran; Vinod Kumar vinodverma507@gmail.com Department of Botany, DAV University, Jalandhar, Punjab 144012, India © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 88 A. KESHAVARZI AND V. KUMAR done the spatial distribution of heavy metals by GIS- which have great influence on the agricultural soil based geostatistical techniques (Ander et al., 2013; quality (Dogra et al., 2019). Keeping all these things Petrik, Thiombane, Albanese, Lima, & De Vivo, in mind, the present study was done to assess the pH, 2018; Tóth, Hermann, Szatmári, & Pásztor, 2016; sand, silt, clay, organic carbon (OC), phosphorus (P), Ungureanu, Iancu, Pintilei, & Chicoș, 2016; Vasu et potassium (K), and heavy metals (Fe, Mn, Cu, and al., 2017; Zou et al., 2015). The objective was to Zn) in agricultural soils of Northeastern Iran. The elucidate the causes for high heavy metal contents multivariate techniques such as Pearson’s correlation, and to recognize the main regions for further heatmap, principal component analysis (PCA), and evaluation. nonmetric multidimensional scaling (NMDS) in However, from the last few years the pollution combination with contamination indices such as CF, status of agricultural soils have been widely studied EF, Igeo, and ecological RI and modified ecological but less data were presented about source apportion- risk index (MRI) were applied to determine the pos- ment and ecological risks of heavy metals in agricul- sible sources, level of contamination, and ecological ture soils of diverse land types in Mesopotamian risks posed by Cu, Zn, and Mn in agricultural soils of countries and mainly in Northeastern Iran. Wheat, Neyshabur plain. Further, the spatial distribution barley, corn, etc. are the major crops of Northeastern maps of EF, Igeo, and MRI were also done to define Iran (Esmaeilzadeh et al., 2018). Khorasan Steel the sites that contain high contamination and ecolo- Complex is the biggest incorporated steelmaking gical risks that required further elucidated explora- plant in Northeastern Iran. In addition to the produc- tion. The outcomes of this study will present a tion of iron and steel, various byproducts, i.e., sludge, baseline data about soil quality status in agricultural oxide layers, slag, metal shell, and dust, are generated soils, which may be helpful in protecting the food and are deposited in surrounding regions of this area crop quality and ultimately human health. Legend ! Sampling points Boundary Elevation (m) 1,141 - 1,242 1,242.000001 - 1,343 1,343.000001 - 1,502 1,502.000001 - 1,778 1,778.000001 - 2,275 lReference: WGS_1984_UTM_Zone_ 40N 03 1.75 .5 7 Kilometers Figure 1. Location of study area and sampling points. GEOLOGY, ECOLOGY, AND LANDSCAPES 89 Legend Geological unit Abbreviation COm Jmz Cl Jph Db K2a.bv E1c Kurl E1l L.E-Ogr E1s Mur E2c Murm E2f Ogr E2m Osh E2sht Pel E3c Pgkc E3m PlQc Ea.bvt PlQdv Eat Plc Eav Psch2 Eavt Pz Jd Qal Jl Qcf Qft1 Qft2 Sn di-gb du pC-C sp1 spr sr tm 01 5 0 20 Kilometers Spatial Reference: WGS_1984_UTM_Zone_40NProjected Coordinate System: WGS_1984_UTM_Zone_40NProjec ted Coordinate System: WGS_1984_UTM_Zone_40NSpatia Figure 2. (a) Geological units map of study area. Materials and method (Bagherzadeh et al., 2016), the soils were classified as Aridisols and Entisols orders (Figure 2(d)). Study area Neyshabur plain, Khorasan-e-Razavi Province, Soil sampling and determination Northeast Iran was chosen for the present study (Figure 1). It is situated between lat. 35°40′N to 36° Using ArcGIS 10.4.1 software, a fishnet sampling 40′N and long. 58°12′E to 59°31′E with a normal design including 300 grids was created (Figure 1) altitude of 1256 m above mean sea level. The climate for an appropriate identification of soil sampling of the studied area is semiarid with an average annual areas to consider the spatial variation of the para- precipitation of 233.7 mm and an average annual meters affecting the agricultural soil quality. The temperature of 14.5°C. The irrigated farming is the grid interval was 500 × 500 m. During soil sam- major land-use approach in the studied area pling, a portable Global Positioning System was (Bagherzadeh, Ghadiri, Darban, & Gholizadeh, applied to exactly find the sampling locations. 2016). The general physiographic trend of the plain Twelve locationsof 300 gridswereurbanized extends in the NW–SE direction. Figure 2(a–c) shows (mostly limited by fence) and were not sampled. A the geological map, land types, and land uses of study total of 68 sampling sites (0–30 cm) were selected area. In the studied area, the major geological unit is and samples were collected in triplicates as compo- Qft2, which describes the low level of Piedmont fan site samples from different sites. Approximate five and valley terrace deposits. The description of all the subsamples of soil were pooled to make the com- geological units is given in Table 1. The major type of posite sample. All the samples were stored in clean land is presented by code 4, which represents polythene bags and transported to the laboratory. Piedmont plain. The irrigated farming and shortgrass After that, samples were air dried, grounded, and rangeland are the dominated land uses in the studied passed with 2 mm sieve for analysis. pH was mea- area. As seen in Figure 2(d), Aridisols is the major sured with the help of digital pH-meter (Thomas, soil type in the studied area. Based on earlier study 1996). Soil textural properties of sand (0.05–2 mm), 90 A. KESHAVARZI AND V. KUMAR Legend Land type Code 01 5 0 20 Kilometers Projected Coordinate System: WGS_1984_UTM_Zone_40N Figure 2. (b) land types map of study area. silt (0.002–0.05 mm), and clay (<0.002 mm) were rates for samples spiked with standards validated measured by the hydrometer method (Gee & that the results are reasonable (Xiao et al., 2013). Bauder, 1986). Soil organic carbon (SOC) was determined by following the method of Walkley Assessment of contamination level in agricultural and Black method (Walkley & Black, 1934). Soil P soils was measured by using the method of Olsen et al. (1954). K was determined by following the method The various CFs and indices were used to different using extraction with 1 M ammonium acetate heavy metals to find the degree of pollution level and (NH OAC) at pH7 (Thomas, 1996). Heavy metals ecological risks posed by heavy metals such as CF, EF, such as Mn, Fe, Zn, and Cu were measured by Igeo, potential ecological risk (RI), modified potential atomic absorption spectrometer. The soil samples ecological risk (MRI), etc. (Kumar et al., 2018; Tian et were digested with aqua-regia method (HNO :HCl al., 2017). These pollution indices may illustrate a in a ratio of 1:3) by following the method of qualitative threshold or target on ecological risk mea- Kumar, Sharma, and Thukral (2016, 2018). The surement of individual heavy metals. digested samples were filtered and diluted with 20 ml double steam distilled water and used for study. The detection limits of instrument were Mn CF (1.0 mg/l), Fe (7.3 mg/l), Zn (1.6 mg/l), and Cu It gives us indication about the anthropogenic inputs of (1.2 mg/l). For quality assurance and quality con- heavy metals in the agricultural soils (Ahmed et al., 2016). trol, the standards and blanks were run after every It was calculated by following the equation given by five samples to check the 95% accuracy of the Hakanson (1980): machine (Arora et al. 2008). The 95–105% recovery GEOLOGY, ECOLOGY, AND LANDSCAPES 91 Legend Land use Irrigated Farming Rainfed Agriculture Shortgrass Rangeland Shrubland 01 5 0 20 Kilometers Projected Coordinate System: WGS_1984_UTM_Zone_40N Figure 2. (c) land uses map of study area. HM HM HM CF ¼ b EF ¼ HM Fe b s Fe where HM and HM are the concentrations of heavy s b where HM and HM are the heavy metal concentra- s b metals in sampling sites and background environ- tions in the samples and background environment, ment. The background values of heavy metals were respectively, whereas Fe and Fe are the iron concen- s b taken from Taylor and McLennan (1995). CF is trations in the samples and background environment, grouped into four grades: (CF < 1; low contamina- respectively. The EF is categorized into seven types tion), (1 < CF ≤ 3; moderate contamination), such as (EF < 1; no enrichment), (1 ≤ EF < 3; less (3 < CF ≤ 6; high contamination), and (CF > 6; enrichment), (3 ≤ EF < 5; moderate enrichment), very high contamination) (Hakanson, 1980). (5 ≤ EF < 10; moderately enrichment), (10 ≤ EF < 25; high enrichment), (25 ≤ EF < 50; very high enrich- ment), and (EF > 50; exceptionally high enrichment) EF (Marrugo-Negrete, Pinedo-Hernández, & Díez, 2017). It presents the enrichment of heavy metals against the content of background heavy metals (Sakram et al., 2015). The heavy metals geochemically differentiating Igeo with elevated content in the ecosystem and not com- Igeo measured the pollution level on the basis of petent of presenting antagonism or synergism toward concentrations of heavy metals. It was determined as the evaluated heavy metals are used as background heavy metals (Chandrasekaran et al., 2015). Fe was HM preferred to background heavy metal and EF was Igeo ¼ log2 1:5 HM computed as 92 A. KESHAVARZI AND V. KUMAR Legend Soil order Aridisols Entisols/Aridisols Inceptisols Rock Outcrops/Entisols Rock Outcrops/Inceptisols 01 5 0 20 Kilometers Projected Coordinate System: WGS_1984_UTM_Zone_40N Figure 2. (d) soils map of study area. where HM and HM are the concentrations of heavy lithogenic additions of heavy metals in agricultural soils, s b metal in samples and background environment, EF replaced for computation of potential ecological RI. respectively. The constant 1.5 signifies changes in The ecological RI computed from EF is called potential concentrations of heavy metals in the environment MPI (Kumar et al. 2018). It is defined as the multi- (Wei & Yang, 2010). On the basis of Igeo values, it is plication of EF and T of individual heavy metals. It grouped as Igeo ≤ 0, no pollution; Igeo (0–1), mod- was determined by the following equation: erate pollution; Igeo (1–2), strong pollution; Igeo (2– MRI ¼ EF T n r 3), high pollution; Igeo (3–4), very high pollution; where EF and T are the EF and toxicological Igeo (4–5), severe pollution; and Igeo ≤5, extreme n r response factor of individual heavy metals, respec- pollution (Loska, Wiechuła, & Korus, 2004). tively. The grades used for risk assessment are as follows: Er < 40 (low risk), 40–80 (moderate risk), Ecological risk assessment (RI and MRI) 80–160 (considerable risk), 160–320 (high risk), and >320 (very high risk). The potential ecological RI was computed to evaluate the ecological risk assessment of heavy metals in the agricultural soils. It is defined as multiplication of CF Geostatistical modeling of each heavy metal and toxicological response factor (T To illustrate the spatial distributions of contamina- ) of individual heavy metals, viz., Cu (5), Zn, and tion indices and factors such as EF, Igeo, and MRI for Mn (1) (Kumar et al., 2018). It was determined by the Mn, Cu, and Zn heavy metals, kriging was accepted following equation: as done by various researchers (Liu et al., 2017; RI ¼ CF T n r Masoud, Koike, Mashaly, & Gergis, 2016). The kri- where CF and T are the CF and toxicological response ging technique is one of the best linear unbiased n r factor of individual heavy metals, respectively. In order techniques that makes satisfactory spatial maps in to determine the ecological risks of anthropogenic and scanty data area and gives stochastic ambiguity of GEOLOGY, ECOLOGY, AND LANDSCAPES 93 error, coefficient of variation, skewness, and kurtosis Table 1. Description of geological units used in the study in PAST software v. 3.15. The data were also analyzed area. Geological for PCA and NMDS. PCA was implemented to assess unit Description the contamination sources in agricultural soils of Cl Dark red medium-grained arkosic to sub-arkosic Northeastern Iran. It was employed by using varimax sandstone and micaceous siltstone COm Dolomite platy and flaggy limestone containing trilobite; rotation with Kaiser Normalization to assess the nor- sandstone and shale malized data after evaluating the compatibility of Db Grey and black, partly nodular limestone with datasets for PCA factors by employing SPSS v. 16 intercalations of calcareous shale di-gb Gabbro to diorite, diorite, and trondhjemite ([IBM, USA] software (Kumar et al., 2018; Zhang et du Dunite al. 2018). In NMDS, grade alterations among the E1c Pale-red, polygenic conglomerate, and sandstone E1l Nummulitic limestone sampling sites in multidimensional space are sus- E1s Sandstone, conglomerate, marl, and sandy limestone tained in 2D or 3D space using correlation as simi- E2c Conglomerate and sandstone E2f Sandstone, calcareous sandstone, and limestone larity measure. Heatmap was prepared by using R E2m Pale red marl, gypsiferous marl, and limestone programming software v. 3.1.3. E2sht Tuffaceous shale and tuff E3c Conglomerate and sandstone E3m Marl, sandstone, and limestone Ea.bvt Andesitic to basaltic volcanic tuff Results and discussion Eat Andesitic tuff Eav Andesitic volcanics The descriptive statistical analysis of pH, sand, silt, Eavt Andesitic volcanic tuff clay, OC, P, K, Fe, Cu, Mn, and Zn is presented in Jd Well-bedded to thin-bedded, greenish-grey argillaceous limestone with intercalations of calcareous shale Table 2. The pH was recorded in the range of 7.5– Jl Light grey, thin-bedded to massive limestone 8.3 in different sampling sites. The slightly alkaline Jmz Grey thick-bedded limestone and dolomite Jph Phyllite, slate, and meta-sandstone nature of pH is responsible for decreasing the mobi- K2a.bv Andesitic and basaltic volcanic rocks lity of heavy metals in the soils (Tian et al., 2017). Kurl Undifferentiated pelagic limestone and radiolarian chert L.E-Ogr Late Eocene – Early Oligocene granite The range of 0.17–0.73% for OC was recorded in the Mur Red marl, gypsiferous marl, sandstone, and present study and affected the retention of heavy conglomerate Murm Light-red to brown marl and gypsiferous marl with metal in the soils (Troeh & Thompson, 2005). The sandstone intercalations P and K were found in the range of 2.4–19.4 mg/kg Ogr Granite and 73.08–261 mg/kg, respectively. Fe content varies Osh Greenish-grey siltstone and shale with intercalations of flaggy limestone from 20,000 to 550,000 mg/kg in worldwide soils pC-C Late Proterozoic–early Cambrian undifferentiated rocks (Bodek,Lyman,&Reehl, 1988) and changes exten- Pel Medium to thick-bedded limestone Pgkc Light-red coarse grained, polygenic conglomerate with sively,evenwithinsameareas becauseofsoiltypes. sandstone intercalations In the present study, Fe content was found in the Plc Polymictic conglomerate and sandstone PlQc Fluvial conglomerate, piedmont conglomerate, and range of 2.31–1.24 mg/kg and their low content was sandstone attributed to sandy texture of agricultural soils. PlQdv Rhyolitic to rhyodacite volcanics Psch2 Metamorphosed turbidite associated with met ultrabasic Normally, Fe content was found low in sandy soils and basic rocks and high in clayey soils (McGovern, 1987). Zn con- Pz Undifferentiated lower Paleozoic rocks tent varies from 10 to 300 mg/kg with a mean value Qal Stream channel, braided channel, and floodplain deposits Qcf Clay flat of 50 mg/kg in worldwide soils (Alloway, 2008). In Qft1 High-level piedmont fan and valley terrace deposits the present study, Zn concentration ranges from Qft2 Low-level piedmont fan and valley terrace deposits Sn Greenish grey, shale, sandstone, sandy lime, coral 0.28 to 2.84 mg/kg and low concentration of Zn limestone, and dolomite may be attributed to sandy soil and low OC in sp1 Spilitespilitic andesite and diabasic tuff spr Submarine, vesicular basalt, locally with pillow structure agricultural soils of this area. The Zn content was in association with radiolarian chert also found low as compared to Indian permissible sr Serpentinite tm Tectonic melange – association of ophiolitic limits of soils, i.e., 300–600 mg/kg (Awashthi, 2000) components, pelagic limestone, radiolarian chert, and and 30 mg/kg (European Union, 2009). The mean shale with or without Eocene sedimentary rocks worldwide upper crustal concentration of Mn is 600 mg/kg and bulk continental crust content is 1400 mg/kg (Taylor and McLennan 1995). The Mn the maps (Burrough & McDonnell, 2015). The kri- content for the present study was found in the range ging interpolations of EF, Igeo, and MRI were com- of 1.64–7.18 mg/kg which is lower than the average puted by employing ArcGis software to show their upper crustal and bulk continental crust concentra- spatial distribution maps (Chen et al., 2016). tions. The Mn content found in the present study was found lesser than limit of 2000 mg/kg given by European Union (2009). The Mn content was also Statistical analysis attributed to low OC and sandy texture of agricul- The data were analyzed for various descriptive statis- tural soils. The range of Cu recorded in the present tical analysis mean, standard deviation, standard study was found lower than the limits of Iran EPA 94 A. KESHAVARZI AND V. KUMAR Table 2. Descriptive statistics of soil properties from agricultural fields of Neyshabur plain. Sites pH Sand (%) Silt (%) Clay (%) OC (%) P (mg/kg) K (mg/kg) Fe (mg/kg) Mn (mg/kg) Zn (mg/kg) Cu (mg/kg) 1 7.7 47.4 29.6 23.0 0.74 7.2 219.4 2.10 4.12 0.44 1.24 2 7.9 31.4 37.6 31.0 0.92 38.4 404.8 3.50 21.06 0.68 1.94 3 7.7 39.4 35.6 25.0 0.71 16.8 387.6 2.08 8.38 0.46 1.68 4 8.0 27.4 37.6 35.0 0.46 8.8 288.8 1.66 4.14 0.34 1.32 5 7.8 41.4 31.6 27.0 0.71 43.6 273.0 2.14 10.26 0.40 1.52 6 7.9 43.4 25.6 31.0 0.57 2.4 219.4 2.14 5.90 0.30 1.32 7 7.8 29.4 41.6 29.0 0.89 9.6 273.0 2.84 18.22 0.54 1.50 8 8.0 37.0 34.0 29.0 0.53 7.6 189.8 1.40 6.42 0.28 1.02 9 7.6 47.0 32.0 21.0 0.43 10.4 168.1 1.70 9.90 0.30 0.84 10 7.5 59.0 24.0 17.0 0.71 36.8 234.4 2.42 4.68 1.88 0.98 11 7.7 55.0 30.0 15.0 0.57 7.6 119.4 3.94 6.18 0.30 0.74 12 7.5 49.0 30.0 21.0 0.85 5.6 133.1 2.66 9.80 0.58 1.60 13 7.7 31.0 40.0 29.0 0.74 7.6 197.1 2.64 6.52 0.86 1.22 14 7.7 35.0 42.0 23.0 1.34 58.8 792.4 3.10 18.08 0.46 1.82 15 7.7 37.0 32.0 31.0 0.67 8.4 175.3 1.90 7.56 0.46 1.40 16 7.8 35.0 32.0 33.0 0.59 14.4 265.2 2.26 11.40 0.50 1.42 17 7.6 25.0 42.0 33.0 0.67 37.2 204.5 3.26 9.28 0.38 1.34 18 7.8 39.0 34.0 27.0 0.74 10.0 147.0 4.22 4.44 0.42 0.90 19 7.8 43.0 34.0 23.0 0.71 24.0 204.5 2.50 10.26 1.16 1.30 20 8.0 43.0 32.0 25.0 0.82 32.4 484.6 1.96 9.76 0.78 1.32 21 7.6 47.0 28.0 25.0 0.37 5.6 249.7 1.86 5.54 0.48 1.46 22 7.8 45.0 32.0 23.0 0.37 6.0 273.0 1.92 3.04 0.50 1.42 23 7.8 25.0 46.0 29.0 0.56 8.8 320.9 2.32 16.44 0.86 1.46 24 7.8 35.0 40.0 25.0 1.00 49.6 439.8 2.58 15.20 0.60 1.66 25 7.7 43.0 34.0 23.0 0.46 20.4 280.9 2.30 10.68 0.66 0.78 26 7.8 34.6 42.4 23.0 0.67 8.4 320.9 1.24 3.04 0.42 0.80 27 7.6 46.6 34.4 19.0 0.74 35.6 404.8 1.94 9.92 0.74 1.34 28 7.7 26.6 48.4 25.0 0.56 20.0 249.7 1.80 8.54 0.98 1.12 29 8.0 40.6 32.4 27.0 0.48 12.8 249.7 1.56 8.12 1.80 0.94 30 7.8 34.6 42.4 23.0 0.84 11.2 370.7 1.34 2.86 0.48 1.14 31 7.8 30.6 43.4 26.0 0.59 19.6 304.8 1.76 7.48 0.36 1.26 32 7.8 34.6 39.4 26.0 0.38 10.4 370.7 2.74 1.64 1.56 0.82 33 8.0 41.6 32.4 26.0 0.76 5.2 312.8 1.60 4.58 1.16 1.22 34 8.1 37.6 41.4 21.0 0.38 4.4 219.4 1.66 2.92 1.22 1.18 35 7.9 43.0 37.4 19.6 0.95 50.4 288.8 1.38 3.62 7.44 0.98 36 8.0 47.0 31.4 21.6 1.60 19.6 616.7 1.32 4.48 2.18 1.08 37 8.0 42.0 36.4 21.6 0.59 10.4 219.4 2.82 4.40 0.70 1.30 38 8.1 34.0 50.4 15.6 1.62 60.0 577.9 1.90 5.86 0.98 1.12 39 7.8 41.0 37.4 21.6 1.22 30.8 448.6 2.08 5.54 1.78 1.02 40 8.0 29.6 42.4 28.0 0.68 7.2 147.0 2.32 4.10 3.08 1.06 41 8.2 40.6 37.4 22.0 0.44 7.6 119.4 1.82 4.16 1.06 0.96 42 8.0 42.6 35.4 22.0 0.72 4.0 168.1 1.90 2.82 1.18 1.08 43 8.1 42.6 41.4 16.0 0.63 10.0 112.6 2.84 4.36 8.56 1.00 44 8.0 47.6 30.4 22.0 0.91 48.0 133.1 2.20 3.84 12.32 1.04 45 8.2 39.6 40.4 20.0 0.61 20.4 119.4 2.26 6.22 7.74 1.18 46 8.0 43.6 40.4 16.0 1.48 48.0 413.5 2.56 4.60 9.22 1.24 47 8.0 40.0 44.0 16.0 0.23 13.2 197.1 2.88 5.26 11.56 1.26 48 8.1 41.0 43.0 16.0 0.74 11.2 189.8 2.46 5.16 5.02 0.82 49 8.2 36.0 44.0 20.0 1.29 5.6 105.9 3.82 4.74 3.26 1.32 50 8.1 27.0 46.0 27.0 0.65 8.4 273.0 3.00 4.64 5.62 0.76 51 8.3 68.0 21.0 11.0 1.45 4.8 73.1 2.98 6.20 10.58 1.10 52 7.6 65.0 25.0 10.0 0.36 63.2 105.9 2.48 8.14 0.82 1.62 53 8.0 54.0 32.0 14.0 0.59 2.4 133.1 2.74 4.18 10.46 1.08 54 8.0 50.0 35.0 15.0 0.49 24.8 304.8 2.64 8.36 1.46 0.96 55 8.1 54.0 31.0 15.0 0.17 5.2 119.4 3.52 4.76 13.94 0.76 56 8.0 54.0 34.0 12.0 0.68 34.4 112.6 2.38 5.16 10.10 1.68 57 8.0 52.0 34.0 14.0 0.72 40.4 112.6 3.00 17.54 13.90 0.72 58 8.0 33.0 45.0 22.0 0.51 33.2 257.4 3.58 4.40 1.54 0.76 59 7.8 60.6 29.0 10.4 0.25 30.4 337.3 2.56 9.88 6.66 0.96 60 7.8 44.6 35.0 20.4 0.89 8.8 140.0 2.14 5.74 1.12 1.34 61 7.9 31.6 40.0 28.4 0.34 4.0 154.0 2.40 7.28 2.70 0.86 62 7.9 39.6 47.0 13.4 0.68 27.2 257.4 2.44 9.58 1.24 0.90 63 8.1 37.6 35.0 27.4 0.44 36.4 211.9 1.38 4.20 0.98 0.80 64 8.2 31.6 41.0 27.4 0.78 4.0 189.8 2.34 6.18 1.66 0.80 65 8.3 25.6 45.0 29.4 1.46 11.2 413.5 1.60 7.34 0.62 1.40 66 8.2 34.6 43.0 22.4 0.97 16.4 379.1 1.74 7.86 1.32 0.96 67 7.9 25.6 45.0 29.4 0.65 6.8 249.7 1.38 6.52 7.84 1.28 68 7.9 23.6 47.0 29.4 1.08 8.8 320.9 1.48 4.72 1.42 0.94 Mean 7.90 40.29 36.99 22.72 0.73 19.45 261.05 2.31 7.18 2.84 1.16 Min 7.5 23.6 21 10 0.17 2.4 73.08 1.24 1.64 0.28 0.72 Max 8.3 68 50.4 35 1.62 63.2 792.4 4.22 21.06 13.94 1.94 SE 0.02 1.18 0.78 0.73 0.04 1.97 16.16 0.08 0.49 0.46 0.04 SD 0.19 9.72 6.41 5.99 0.32 16.25 133.27 0.68 4.06 3.78 0.29 Skewness −0.01 0.57 −0.13 −0.26 1.07 1.09 1.42 0.66 1.61 1.68 0.47 Kurtosis −0.59 0.34 −0.53 −0.61 1.05 0.13 3.10 0.20 2.53 1.58 0.34 CV 2.43 24.13 17.32 26.35 44.47 83.51 51.05 29.23 56.51 133.05 24.81 GEOLOGY, ECOLOGY, AND LANDSCAPES 95 Fe 0.8 Mn 0.6 Zn 0.4 Cu 0.2 pH OC 0 Sand -0.2 Silt -0.4 Clay -0.6 AvaP -0.8 AvaK -1 Figure 3. Pearson correlation analysis of soil properties and heavy metals from Neyshabur plain (N = 68). Figure 4. Heatmap of sampling sites and soil properties. Table 3. Principal component analysis of soil properties from Neyshabur plain. Initial Eigen values Extraction sums of squared loadings Rotation sums of squared loadings Components Total Var (%) Cumulative (%) Total Var (%) Cumulative (%) Total Var (%) Cumulative (%) 1 2.91 26.5 26.5 2.91 26.5 26.5 2.51 22.8 22.8 2 2.18 19.8 46.4 2.18 19.8 46.4 2.10 19.1 41.9 3 1.76 16.0 62.4 1.76 16.0 62.4 2.06 18.7 60.7 4 1.21 11.0 73.4 1.21 11.0 73.4 1.40 12.7 73.4 Component matrix Rotated component matrix Variables PC1 PC2 PC3 PC4 PC1 PC2 PC3 PC4 pH −0.187 −0.446 0.683 0.043 pH 0.190 0.811 0.093 −0.009 Sand −0.812 0.484 −0.114 −0.261 Sand −0.983 0.047 −0.045 0.063 Silt 0.551 −0.394 0.498 0.309 Silt 0.800 0.313 0.231 0.106 Clay 0.729 −0.365 −0.349 0.093 Clay 0.740 −0.412 −0.174 −0.216 OC 0.322 0.306 0.634 −0.224 OC 0.098 0.167 0.780 −0.062 P 0.115 0.722 0.347 −0.072 P −0.265 −0.153 0.717 0.229 K 0.657 0.366 0.329 −0.316 K 0.256 −0.244 0.776 −0.216 0.265 0.051 0.771 Fe −0.103 0.056 −0.132 0.851 Fe −0.299 Mn 0.384 0.589 −0.083 0.481 Mn 0.143 −0.521 0.323 0.580 Zn −0.646 0.019 0.483 0.223 Zn −0.373 0.647 0.007 0.377 Cu 0.431 0.548 −0.201 0.084 Cu 0.052 −0.615 0.345 0.183 PC: Principal components; bold numbers represent significant loadings. guidelines (100 mg/kg) and Earth’scrust (26 mg/kg) Vega,Silva,&Andrade, 2014; Cerqueira, Vega, (Department of Environmental Protection, 2017; Silva, & Andrade, 2012). Mirsal, 2008). The skewness and kurtosis of pH, Pearson’s correlation analysis was applied to soil sand, silt, clay, Fe, and Cu were found less than properties to evaluate the relationship among differ- oneand arenormallydistributed (Beaver, Beaver, ent soil properties (Figure 3). pH showed negative &Mendenhall, 2012). The skewness and kurtosis of correlation with Mn and Cu, and positive correlation OC, K, Mn, and Zn were found greater than one and with Zn. The sand content was positively associated indicate right-handed skewness and leptokurtic with Zn. Clay content showed negative relationship (Beaver et al., 2012). The coefficient of variation of with the Zn content. The P showed positive correla- heavy metals such as Zn, Mn, Fe, and Cu showed tion with K and Mn. Mn and Cu are positively great variations representing anthropogenic activ- correlated with each other. The high correlations of ities have great influence on the heavy metals con- heavy metals among each other indicated their simi- tent (Cai et al. 2015). Some soil properties show high lar source that may be due to natural as well as degree of standard deviation which may be due to human activities. The parent rock material was the the lack of uniformity of heavy metal distribution in major natural sources because these metals are agricultural soils of Northeastern Iran (Arenas-Lago, mainly components of Earth’s crust (Taylor & Fe Mn Zn Cu pH OC Sand Silt Clay AvaP AvaK 96 A. KESHAVARZI AND V. KUMAR 0.03 0.02 68 23 67 65 61 64 26 30 40 33 0. 34 0121 8 22 36 49 66 42 15 1 16 37 29 53 12 60 47 28 -0.4 -0.3 -0.2 -0.1 0.1 0.2 0.3 0.4 0.5 51 45 62 54 -0.01 63 14 -0. 1702 5 -0.03 -0.04 -0.05 -0.06 Coordinate 1 0 300 600 9001200 1500 1800 2100 2400 2700 Target rank (b) Figure 5. (a) NMDS scatter plot (95% eclipse) of different agricultural sites from Neyshabur plain on the basis of soil properties using correlation as a similarity measure and (b) NMDS Shepard 2-D plot of different agricultural sites from Neyshabur plain on the basis of soil properties (stress = 0.016, R for axis 1 = 0.997, axis 2 = 0.01001). McLennan, 1995). The irrigated polluted water by group which may be attributed to similarities in soil discharge of sewage and industrial effluents was the properties of these sites. major anthropogenic source (Nagajyoti, Lee, & PCA was also employed to soil properties and Sreekanth, 2010; Tian et al., 2017). heavy metals (Table 3). The first four PC accounted Heatmap was employed to both soil properties and for 73.4% of variation with Eigenvalues found sampling sites (Figure 4). The heatmap indicates that greater than one. The loading of PCA in compo- Zn, Fe, Cu, Mn, and OC are included in the same nent matrix explained that PC1 was contributed by group. K formed the different group. Sand, silt, and soil textural properties (sand, silt, and clay), K, and clay formed the same group. Heatmap of sampling Zn and explained 26.5% of total variation. The high sites showed that sites 20, 24, 39 65, 2, 27, 46, 3, 30, coefficient of variation of K and Zn indicated that 32, and 66 are included in the same cluster. The sites anthropogenic activities have great affect on these 9, 42, 15, 18, 40, 61, 11, 55, 12, 60, 53, 41, 43, 49, 45, parameters. The human activities such as smelting 44, 56, 57, 52, and 51 are also included in the same and industrial discharge affected the Zn content Coordinate 2 Obta ined rank GEOLOGY, ECOLOGY, AND LANDSCAPES 97 Figure 6. (a–e) Box plots of different heavy metals for (a) contamination factor, (b) enrichment factor, (c) geoaccumulation index, (d) potential ecological risk of heavy metals (Er), and (e) modified potential ecological risk of heavy metals (mEr). (Singh & Kumar, 2017). P, Mn, and Cu contribute showed four points such as point no. 14, point no. to PC2 (19.8%). pH and OC contributed to PC3 36, point no. 38, and point no. 52 separated from (16%). PC4 (11%) was contributed by Fe. After the main group (Figure 5(a)). Because the stress in varimax rotation with Kaiser normalization, PC1 NMDS Shepard curve (Figure 5(b)) is 0.016 less (22.8%) accounted for textural properties. The pH, than 0.05, the statistics demonstrated a good fitto Zn, and Cu are contributed to PC2 (19.1%). PC3 NMDS data (Kaur et al., 2018). (18.7%) was contributed by OC, P, and K. Fe and Mn are contributed by PC4 (12.7%). The average Assessment of heavy metal pollution and values of Fe, Mn, Zn, and Cu were found less than ecological risks the permissible limits of worldwide soils, Indian limits for soil, Iran EPA guidelines, and Earth’s The anthropogenic effect of heavy metals was assessed crust and revealed that natural sources mainly by using CF (Figure 6(a)). The CF of all Zn, Cu, and Mn affected their concentration in addition to the was found less than one, indicating low contamination anthropogenic activities (Zhang et al. 2018). The of these heavy metals in the agricultural soils. EF was Co-efficient of variation (CV) of Fe and Cu was also computed to determine the anthropogenic and found low and belong to low spatial unevenness lithogenic effects of Cu,Mn, andZninthe agricultural grade. Due to this, both Fe and Cu may be origi- soils (Figure 6(b)). The results of EF showed that 99% nated from natural sources. Ke et al. (2017) and agricultural samples were highly enriched with Cu, Zn, Liang et al. (2017) also found such type of findings and Mn. The results further showed that each heavy in their studies. The results of NDMS scatter plot metal showed very high enrichment in the studied area. 98 A. KESHAVARZI AND V. KUMAR Legend Prediction Map Mn (EF) 34.2387132 – 70.8275663 70.8275663 – 100.194776 100.194776 – 123.765698 123.765698 – 142.684362 142.684362 – 166.255285 166.255285 – 195.622494 195.622494 – 232.211347 232.211347 – 277.797707 277.797707 – 334.594134 334.594134 – 405.35728 Boundary 03 1.75 .5 7 Kilometers Legend Prediction Map Zn (EF) 33.1271895 – 69.1497722 69.1497722 – 91.8966643 91.8966643 – 127.919247 127.919247 – 184.965566 184.965566 – 275.305631 275.305631 – 418.370559 418.370559 – 644.931978 644.931978 – 1,003.72066 1,003.72066 – 1,571.90795 1,571.90795 – 2,471.70443 Boundary 03 1.75 .5 7 Kilometers Figure 7. (a–i) Spatial distribution map of Mn, Zn, and Cu (8a–c; EF), Mn, Zn, and Cu (8d–f; Igeo), and Mn, Zn, and Cu (8g–i; MRI) using Kriging interpolation method. The Igeo was also calculated for heavy metals (Figure 6 (Figure 6(e)). From the results of RI, it was found that (c)). The results of Igeo showed that 86.7% agricultural Er values for heavy metals were found less than 40, samples were highly polluted with Cu, Zn, and Mn. The indicating low ecological risks of these heavy metals. Igeo of Mn for all the sites showed extreme pollution, The results of MRI revealed that about 52.4% agri- whereas Zn and Cu showed (92.6% and 66.1%, respec- cultural samples showed very high ecological risks of tively) very high pollution in the studied area. heavy metals, i.e., Cu, Zn, and Mn. The mEr values of The ecological risks of Cu, Mn, and Zn were Mn showed that approximate 48.5% agricultural sam- evaluated by potential ecological RI (Figure 6(d)) ples showed high-to-very high ecological risks. The and modified potential ecological risk index (MRI) mEr values of Zn showed that 69.1% samples showed GEOLOGY, ECOLOGY, AND LANDSCAPES 99 Legend Prediction Map Cu (EF) 232.067005 – 338.632564 338.632564 – 437.009806 437.009806 – 527.827907 527.827907 – 611.667701 611.667701 – 689.065387 689.065387 – 766.463073 766.463073 – 850.302866 850.302866 – 941.120968 941.120968 – 1,039.49821 1,039.49821 – 1,146.06377 Boundary 0 1.75 3.5 7 Kilometers Legend Prediction Map Mn (Igeo) 177.728109 – 330.788741 330.788741 – 420.01414 420.01414 – 472.027331 472.027331 – 502.347981 502.347981 – 554.361172 554.361172 – 643.586571 643.586571 – 796.647203 796.647203 – 1,059.21321 1,059.21321 – 1,509.62889 1,509.62889 – 2,282.28902 Boundary 03 1.75 .5 7 Kilometers Figure 7. (Continued) considerable ecological risks, whereas all the sampling Igeo (Figure 7(d–f)), and MRI (Figure 7(g–i)) (Liu sites for mEr of Cu showed very high ecological risks. et al., 2017). Ten groups were used for each para- meter on the basis of their original content. The geostatistical analysis was not employed to CF and Spatial distribution of soil properties, heavy RI because values of these indices showed low con- metals, and EF, Igeo, and MRI tamination and ecological risks. The spatial distribu- Geostatistical analysis presents an equitable determi- tion maps of Zn for EF and Igeo showed that it was nation of different parameters at sites without sam- accumulated more in central regions as compared to pling. Plotting the spatial distribution of agricultural Southern and Northeastern regions of the studied soils is important in estimating the threats of con- area. The spatial maps of Mn and Cu for EF showed tamination and ecological risks at diverse sites. The that these heavy metals were predominately accu- spatial interpolation technique such as kriging mulated in Southern regions of the studied area. assessment has been mainly employed in enlighten- Similarly, spatial maps of Mn and Zn for Igeo ing the spatial distribution of EF (Figure 7(a–c)), observed that these are distributed more in 100 A. KESHAVARZI AND V. KUMAR Legend Prediction Map Zn (Igeo) 3.98965088 – 5.64543725 5.64543725 – 6.59156275 6.59156275 – 8.24734912 8.24734912 – 11.1450921 11.1450921 – 16.2163467 16.2163467 – 25.0914 25.0914 – 40.6233694 40.6233694 – 67.8054115 67.8054115 – 115.375903 115.375903 – 198.627619 Boundary 0 1.75 3.5 7 Kilometers Legend Prediction Map Cu (Igeo) 3.61235995 – 4.11193157 4.11193157 – 4.54627562 4.54627562 – 4.92390866 4.92390866 – 5.35825271 5.35825271 – 5.85782433 5.85782433 – 6.43241908 6.43241908 – 7.09330354 7.09330354 – 7.85343626 7.85343626 – 8.72772185 8.72772185 – 9.73330319 Boundary 03 1.75 .5 7 Kilometers Figure 7. (Continued) Northeastern regions and their distribution in cen- great influence on the soil properties in addition to the tral regionsshowedlessaccumulation. Spatial dis- anthropogenic sources. The results of CF and RI indi- tribution map of Cu for MRI showed highest cated that heavy metals posed low contamination and distribution toward Southern and Northwestern ecological risks in the agricultural soils. The results of regions, whereas in central part of the study area, EF for Cu, Zn, and Mn revealed very high enrichment, it has a lower value. Similarly, Mn also showed whereas Igeo values of Zn (92.6%) and Cu (66.1%) similar trend for MRI values. These results are also showed very high pollution, and all values of Igeo for in associations with heatmap analysis where both Mn posed extreme pollution in agricultural soil sam- heavy metals are included in the same cluster. The ples of studied area. The mEr values for Zn and Mn spatial distribution of Zn was more in central part, showed that 69.1% and 48.5% agricultural soil samples whereas lower values were observed in South and showed mEr values >320 and posed very high ecolo- Northwestern regions of the studied area. gical risks, whereas mEr values of Cu for all the sam- ples showed very high ecological risk. The results of geostatistcal analysis revealed that spatial distribution Conclusions maps of Zn for EF, Igeo, and MRI were distributed The average concentrations of Fe, Mn, Cu, and Zn more in Central regions, whereas spatial distribution were recorded lower than worldwide soils, Indian lim- of Cu and Mn for these indices was more in South and its for soil, Iran EPA guidelines, and Earth’s crust. Northeastern regions of the studied area. Further stu- Heatmap and PCA showed that natural sources have dies are needed to understand the sources that are GEOLOGY, ECOLOGY, AND LANDSCAPES 101 Legend Prediction Map Mn (MRI) 34.2387132 – 70.8275663 70.8275663 – 100.194776 100.194776 – 123.765698 123.765698 – 142.684362 142.684362 – 166.255285 166.255285 – 195.622494 195.622494 – 232.211347 232.211347 – 277.797707 277.797707 – 334.594134 334.594134 – 405.35728 Boundary 0 1.75 3.5 7 Kilometers Legend Prediction Map Zn (MRI) 33.1271895 – 69.1497722 69.1497722 – 91.8966643 91.8966643 – 127.919247 127.919247 – 184.965566 184.965566 – 275.305631 275.305631 – 418.370559 418.370559 – 644.931978 644.931978 – 1,003.72066 1,003.72066 – 1,571.90795 1,571.90795 – 2,471.70443 Boundary 03 1.75 .5 7 Kilometers Legend Prediction Map Cu (MRI) 1,160.33503 – 1,693.16282 1,693.16282 – 2,185.04903 2,185.04903 – 2,639.13954 2,639.13954 – 3,058.3385 3,058.3385 – 3,445.32693 3,445.32693 – 3,832.31536 3,832.31536 – 4,251.51433 4,251.51433 – 4,705.60484 4,705.60484 – 5,197.49105 5,197.49105 – 5,730.31884 Boundary 0 1.75 3.5 7 Kilometers Figure 7. (Continued) 102 A. KESHAVARZI AND V. KUMAR responsible for distribution of heavy metals. The data bank from a copper mine. Science of the Total Environment, 421, 220–229. of this study may help the local government managers Chandrasekaran, A., Ravisankar, R., Harikrishnan, N., to ascertain the pollution reduction approaches and to Satapathy, K. K., Prasad, M. V. R., & Kanagasabapathy, K. protect food safety. V. (2015). Multivariate statistical analysis of heavy metal concentrationinsoils of Yelagiri Hills,Tamilnadu,India – Spectroscopical approach. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 137, 589–600. Disclosure statement Chen, H., Teng, Y., Lu, S., Wang, Y., Wu, J., & Wang, J. No potential conflict of interest was reported by the (2016). Source apportionment and health risk assess- authors. ment of trace metals in surface soils of Beijing metropo- litan, China. Chemosphere, 144, 1002–1011. Department of Environmental Protection. 2017. Islamic Republic of Iran. Retrieved from https://www.doe.ir/ ORCID Portal/Home Dogra, N., Sharma, M., Sharma, A., Keshavarzi, A., Ali Keshavarzi http://orcid.org/0000-0003-3330-6500 Minakshi, Bhardwaj, R., & Kumar, V (2019). Pollution assessment and spatial distribution of roadside agricul- tural soils: A case study from india. International Journal References of Environmental Health Research, 1-14. doi:10.1080/ Ahmed, F., Fakhruddin, A. N. M., Imam, M. T., Khan, N., 09603123.2019.1578865 Khan,T.A.,Rahman,M.M.,&Abdullah,A.T.M. (2016). Esmaeilzadeh, M., Jaafari, J., Mohammadi, A. A., Spatial distribution and source identification of heavy Panahandeh, M., Javid, A., & Javan, S. (2018). metal pollution in roadside surface soil: A study of Dhaka Investigation of the extent of contamination of heavy Aricha highway, Bangladesh. Ecological Processes, 5(1), 2. metals in agricultural soil using statistical analyses and doi:10.1186/s13717-016-0045-5 contamination indices. Human and Ecological Risk Alloway, B. J. (2008). Zinc in soils and crop nutrition, 2nd. Assessment: an International Journal,1–12. doi:10.1080/ Brussels and Paris: IZA and IFA. 10807039.2018.1460798 Ander, E. L., Johnson, C. C., Cave, M. R., Palumbo-Roe, B., European Union. 2009. Heavy metals in wastes. European Nathanail, C. P., & Lark, R. M. (2013). Methodology for Commission on Environment. Retrieved from http://ec. the determination of normal background concentrations europa.eu/environment/waste/mining/studies/pdf/heavy of contaminants in English soil. Science of the Total metalsreport.pdf Environment, 454–455, 604–618. Facchinelli, A., Sacchi, E., & Mallen, L. (2001). Multivariate Arenas-Lago, D., Vega, F. A., Silva, L. F. O., & Andrade, M. statistical and GIS-based approach to identify heavy metals L. (2014). Copper distribution in surface and subsurface sources in soils. Environmental Pollution, 114,313–324. soil horizons. Environmental Science and Pollution Gee, G. W., & Bauder, J. W. (1986). Particle size analysis. Research, 21(18), 10997-11008. In A. Klute (Ed.), Methods of soil analysis: Part 1. Arora, M., Kiran, B., Rani, S., Rani, A., Kaur, B., & Mittal, Agronomy handbook 9 (pp. 383–411). Madison, WI: N. (2008). Heavy metal accumulation in vegetables irri- American Society of Agronomy and Soil Science gated with water from different sources. Food Chemistry, Society of America. 111(4), 811–815. Hakanson, L. (1980). An Ecological Risk Index for Aquatic Awashthi, S. K. (2000). Prevention of food adulteration act Pollution Control. Water Research, 14(8), 975-1001. no. 37, of 1954, central and state rules as amended for doi:10.1016/0043-1354(80)90143-8. 1999, Editions. New Delhi: Ashoka Law House. Hu, Y., Liu, X., Bai, J., Shih, K., Zeng, E. Y., & Cheng, H. Bagherzadeh, A., Ghadiri, E., Darban, A. R. S., & (2013). Assessing heavy metal pollution in the surface soils Gholizadeh, A. (2016). Land suitability modeling by of a region that had undergone three decades of intense parametric-based neural networks and fuzzy methods industrialization and urbanization. Environmental Science for soybean production in a semi-arid region. Modeling and Pollution Research, 20,6150–6159. Earth Systems and Environment, 2(2), 104. Kaur, R., Kaur, R., Sharma, A., Kumar, V., Sharma, M., Beaver, M. B., Beaver, J. R., & Mendenhall, W. (2012). Bhardwaj, R., & Thukral, A. K. (2018). Microbial pro- Introduction to probability and statistics. New Delhi, duction of dicarboxylic acids from edible plants and India: Cengage Learning. milk using GC-MS. Journal of Analytical Science and Bodek, I., Lyman, W. J., & Reehl, W. F. (1988). In B. T. Technology, 9(1), 21. Walton & R. A. Conway (Eds.), Environmental inorganic Ke, X., Gui, S., Huang, H., Zhang, H., Wang, C., & Guo, W. chemistry: Properties, processes, and estimation methods. (2017). Ecological risk assessment and source identifica- SETAC special publication series. New York: Pergamon. tion for heavy metals in surface sediment from the Burrough, P. A., & McDonnell, R. A. (2015). Principles of Liaohe River protected area, China. Chemosphere, 175, geographical information systems. Oxford: Oxford 473–481. University Press. Kelepertzis, E. (2014). Accumulation of heavy metals in Cai, L., Xu, Z., Bao, P., He, M., Dou, L., Chen, L., & Zhu, Y. agricultural soils of Mediterranean: Insights from G. (2015). Multivariate and geostatistical analyses of the Argolida basin, Peloponnese, Greece. Geoderma, 221, spatial distribution and source of arsenic and heavy 82–90. metals in the agricultural soils in Shunde, Southeast Keshavarzi, A., & Kumar, V. (2018). Ecological risk assess- China. J Geochem Explor, 148, 189–195. ment and source apportionment of heavy metal contam- Cerqueira, B., Vega, F. A., Silva, L. F., & Andrade, L. ination in agricultural soils of Northeastern Iran. (2012). Effects of vegetation on chemical and mineralo- International Journal of Environmental Health Research, gical characteristics of soils developed on a decantation 1–17. doi:10.1080/09603123.2018.1555638 GEOLOGY, ECOLOGY, AND LANDSCAPES 103 Kumar, V., Sharma, A., Kaur, P., Sidhu, G. P. S., Bali, A. S., extraction with sodium bicarbonate. Washington. D.C: Bhardwaj, R., & Cerda, A. (2019). Pollution assessment of Government Printing Office. heavy metals in soils of India and ecological risk assess- Petrik, A., Thiombane, M., Albanese, S., Lima, A., & De ment: A state-of-the-art. Chemosphere.doi:10.1016/j. Vivo, B. (2018). Source patterns of Zn, Pb, Cr and Ni chemosphere.2018.10.066 potentially toxic elements (PTEs) through a composi- Kumar, V., Sharma, A., Minakshi, Bhardwaj, R., & Thukral, tional discrimination analysis: A case study on the A. K. (2018). Temporal distribution, source apportion- Campanian topsoil data. Geoderma, 331,87–99. ment, and pollution assessment of metals in the sedi- Sakram, G., Machender, G., Dhakate, R., Saxena, P. R., & ments of Beas river, India. Human and Ecological Risk Prasad, M. D. (2015). Assessment of trace elements in Assessment: an International Journal,1–20. soils around Zaheerabad Town, Medak District, Andhra Kumar, V., Sharma, A., & Thukral, A. K. (2016). Assessment Pradesh, India. Environmental Earth Sciences, 73, 4511– of soil enzyme activities based on soil samples from the 4524. Beas river bed, India using multivariate techniques. Singh, U. K., & Kumar, B. (2017). Pathways of heavy metals Malaysian Journal of Soil Science, 20,135–145. contamination and associated human health risk in Ajay Liang, J., Feng, C., Zeng, G., Gao, X., Zhong, M., Li, X., . . . River basin, India. Chemosphere, 174, 183–199. Fang, Y. (2017). Spatial distribution and source identifi- Taylor, S. R., & McLennan, S. M. (1995). The geochemical cation of heavy metals in surface soils in a typical coal evolution of the continental crust. Reviews of Geophysics, mine city, Lianyuan, China. Environmental Pollution, 33(2), 241–265. 225, 681–690. Thomas, G. W. (1996). Soil pH and soil acidity. In A. L. Page Liu, H., Zhang, Y., Zhou, X., You, X., Shi, Y., & Xu, J. (Ed.), Methods of soil analysis: Part 2. Agronomy handbook (2017). Source identification and spatial distribution of 9 (pp. 475–490). Madison, WI: American Society of heavy metals in tobacco-growing soils in Shandong pro- Agronomy and Soil Science Society of America. vince of China with multivariate and geostatistical ana- Tian, K., Huang, B., Xing, Z., & Hu, W. (2017). lysis. Environmental Science and Pollution Research, 24 Geochemical baseline establishment and ecological risk (6), 5964–5975. evaluation of heavy metals in greenhouse soils from Loska, K., Wiechuła, D., & Korus, I. (2004). Metal contam- Dongtai, China. Ecological Indicators, 72, 510–520. ination of farming soils affected by industry. Tóth, G., Hermann, T., Szatmári, G., & Pásztor, L. (2016). Environment International, 30(2), 159–165. Maps of heavy metals in the soils of the European Union Marrugo-Negrete, J., Pinedo-Hernández, J., & Díez, S. and proposed priority areas for detailed assessment. (2017). Assessment of heavy metal pollution, spatial dis- Science of the Total Environment, 565, 1054–1062. tribution and origin in agricultural soils along the Sinú Troeh, F. R., & Thompson, L. M. (2005). Soil and soil River Basin, Colombia. Environmental Research, 154, fertility (6th ed.). New Delhi: Wiley. 380–388. Ungureanu, T., Iancu, G. O., Pintilei, M., & Chicoș,M.M. Masoud, A. A., El-Horiny, M. M., Atwia, M. G., Gemail, K. (2016). Spatial distribution and geochemistry of heavy S., & Koike, K. (2018). Assessment of groundwater and metals in soils: A case study from the NE area of Vaslui soil quality degradation using multivariate and geosta- county, Romania. Journal of Geochemical Exploration, tistical analyses, Dakhla Oasis, Egypt. Journal of African 176,20–32. Earth Sciences, 142,64–81. Vasu, D., Singh, S. K., Sahu, N., Tiwary, P., Chandran, P., Masoud, A. A., Koike, K., Mashaly, H. A., & Gergis, F. Duraisami, V. P., . . . Kalaiselvi, B. (2017). Assessment of (2016). Spatio-temporal trends and change factors of spatial variability of soil properties using geospatial tech- groundwater quality in an arid area with peat rich aqui- niques for farm level nutrient management. Soil and fers: Emergence of water environmental problems in Tillage Research, 169,25–34. Tanta District, Egypt. Journal of Arid Environments, Walkley, A., & Black, I. A. (1934). An examination of the 124, 360–376. Degtjareff method for determining soil organic matter McGovern, E. C. (1987). Background concentrations of 20 and a proposed modification of the chromic acid titra- elements in soils with special regard for New York state. tion method. Soil Science, 37,29–38. Albany, NY: Wildlife Pathology Unit, New York State Wei, B., & Yang, L. (2010). A review of heavy metal con- Department of Environmental Conservation. taminations in urban soils, urban road dusts and agri- Mirsal, I. (2008). Soil pollution: Origin, monitoring & reme- cultural soils from China. Microchemical Journal, 94(2), diation. Berlin: Springer-Verlag. 99–107. Mohammadi, A., Hajizadeh, Y., Taghipour, H., Mosleh Xiao, R., Bai, J., Huang, L., Zhang, H., Cui, B., & Liu, X. Arani, A., Mokhtari, M., & Fallahzadeh, H. (2018). (2013). Distribution and pollution, toxicity and risk Assessment of metals in agricultural soil of surrounding assessment of heavy metals in sediments from urban areas of Urmia Lake, northwest Iran: A preliminary and rural rivers of the Pearl River delta in southern ecological risk assessment and source identification. China. Ecotoxicology, 22(10), 1564–1575. Human and Ecological Risk Assessment: an Zhang, J., Li, H., Zhou, Y., Dou, L., Cai, L., Mo, L., & You, International Journal,1–18. doi:10.1080/ J. (2018). Bioavailability and soil-to-crop transfer of 10807039.2018.1438173 heavy metals in farmland soils: A case study in the Nagajyoti, P. C., Lee, K. D., & Sreekanth, T. V. M. (2010). Pearl River Delta, South China. Environmental Heavy metals, occurrence and toxicity for plants: A Pollution, 235, 710–719. review. Environmental Chemistry Letters, 8, 199–216. Zou, J., Dai, W., Gong, S., & Ma, Z. (2015). Analysis of Olsen, S. R., Cole, C. V., Watanabe, F. S., & Dean, L. A. spatial variations and sources of heavy metals in farm- (1954). Estimation of available phosphorus in soils by land soils of Beijing suburbs. PLoS One, 10(2), e0118082.
Journal
Geology Ecology and Landscapes – Taylor & Francis
Published: Apr 2, 2020
Keywords: Agricultural soils; multivariate techniques; heatmap; kriging; ecological risk assessment