Effects of copper mining on heavy metal contamination in a rice agrosystem in the Xiaojiang River Basin,southwest China

Wanqiu Pu·Jiaqi Sun·Fangfang Zhang,3·Xingyue Wen·Wenhu Liu·Chengmin Huang

Abstract Metallic ore mining causes heavy metal pollution worldwide.However,the fate of heavy metals in agrosystems with long-term contamination has been poorly studied.Dongchuan District(Yunnan,southwest China),located at the middle reaches of the Xiaojiang River,is a well-known 2000-year-old copper mining site. In this work,a survey on soil heavy metal contents was conducted using a handheld X-ray fluorescence instrument to understand the general contamination of heavy metals in the Xiaojiang River Basin.Furthermore,river water,soil,and rice samples at six sites along the fluvial/alluvial fans of the river were collected and analyzed to implement an environmental assessment and an evaluation of irrigated agrosystem.V,Zn,and Cu soil levels(1724,1047,and 696 mg·kg-1, respectively) far exceeded background levels.The geo-accumulation indexes(Igeo)showed that cultivated soils near the mining sites were polluted by Cd and Cu,followed by Zn,V,Pb,Cr,Ni,and U.The pollution index(Pi)indicated that rice in the area was heavily polluted with Pb,Cr,Cd,Ni,Zn,and Cu.The difference in orders of metal concentrations between the soil and rice heavy metal contamination was related to the proportion of bioavailable heavy metals in the soil.The crop consumption risk assessment showed that the hazard quotient exceeded the safe threshold,indicating a potential carcinogenic risk to consumers.The Nemerow integrated pollution index and health index indicated that the middle of the river(near the mining area)was the heaviest polluted site.

Keywords Heavy metal·Rice agrosystem·Copper mining·Pollution assessment

1 Introduction

Heavy metals are one of the most problematic types of pollutants due to their wide distribution,persistent toxicity,and bioaccumulation in food chains(Dudka and Adriano 1997；Kabata-Pendias and Mukherjee 2007；Zhang et al.2012；Waterlot et al.2013).Mining activities lead to considerable heavy metal soil contamination worldwide(Zhuang et al.2009；Li et al.2014；Pan et al.2016).Mining activities can release heavy metals into the surrounding environment(Navarro et al.2008；Xu et al.2013；Halim et al.2015),including soil,water,and air,as well as plants and food(Qu et al.2012),and accumulated heavy metals are eventually absorbed into human bodies via food or direct ingestion,creating a threat to human health(Goyer 1996).

Mining wastes that have been discharged down-slope of the river may be used in irrigation water for agricultural fields(Liao et al.2016).Heavy metals can migrate into and accumulate in soil and plants due to long-term or intensive irrigation with wastewater around mining areas(Sikka et al.2009；Bai et al.2011；Marrugo-Negrete et al.2017).However,the distribution and systematic linkages of heavy metals among soil,water,and plants(or crops),especially under high background levels of heavy metals,have been poorly studied(Avkopashvili et al.2017；Zang et al.2017).

China is one of the largest global producers and consumers of metals/metalloids and has 171 mineral varieties with 12%of global total mineral resources(Hu et al.2010；Li et al.2014).In Yunnan Province,southwest China,142 kinds of mineral deposits,which account for 83%of those discovered nationally,were discovered by 2010(CMIAEP 2011；Shi 2013).Copper mining in Dongchuan,Yunnan,which has experienced a long history of mining,can be traced back to the fourth century B.C.,and the proven copper reserves area contained as much as 3,914,000 tons of copper(Gao et al.2012).In addition,local copper-associated ores include numerous metals,e.g.,Pb,Zn,Ag,Cd,Co,and Ni(Xue 1995；Mwesigye et al.2016).Large quantities of mine waste and tailings have been produced over 2000 years(Yang 2004),and several cases of diseases related to environmental pollution were reported among residents in recent years(Xie 2013；Huang et al.2017).Nevertheless,the impacts of copper mining on the environment,particularly river water,soils,and crops,and subsequent human health implications around mining areas are still unclear.

In this study,to better understand the effect of copper mining on soil-rice-agrosystem on the fluvial/alluvial fans along the Xiaojiang River,Dongchuan,southwest China,fast on-site screening and sequential extraction procedure techniques associated with the environmental risk assessment methods on heavy metals in soil,water,and rice samples were executed.Using a handheld XRF analyzer,the heavy metal concentrations of soils and sediments at nearly 300 sites from 12 plots along the Xiaojiang River were detected in the field.Based on the field survey,we systematically determined 9 heavy metals(Cu,Zn,Pb,Cd,Cr,V,Co,Ni,and U)with an inductively coupled plasma atomic emission spectrometer(ICP-AES)in river water,surface soils,and rice(including grains,stems,roots and leaves)at six selected sites located in the lower reach of the Xiaojiang River,Dongchuan.The objectives of this study were to(1)measure the levels of heavy metals in agricultural soil,water,and rice near the mining area,(2)investigate the effect of copper mining on heavy metals in a rice agrosystem,and(3)assess the ecological risk of soil and rice polluted with heavy metals.

2 Materials and methods

2.1 Study area

This study was carried out in the reaches of the Xiaojiang River Basin (25°37′～27°00′N,103°08′～103°06′E),located in Dongchuan,Yunnan,southwest China(Fig.1).Rainfall in this area is highly seasonal,with a dry period between November and April and a rainy season between March and October.The mean annual precipitation and evaporation are ～700 mm and ～3700 mm, respectively.The average annual temperature is approximately 20°C.The Xiaojiang River drainage covers an area of 3086 km2.Despite the shortage of arable land due to the mountains and hills in Dongchuan,the flat land on the fluvial/alluvial fans along the Xiaojiang River has been cultivated for agricultural use(Fig.1).These croplands are irrigated with the river water.Paddy rice is the most common crop cultivated in these soils on the flood fans.With long-term irrigation of river water with a high concentration of sediment that could be as high as 220 kg·m-3(Cheng et al.2000；Hou 2006),the soils were gradually silted up and thickened due to the sediment deposition.

Deposit evacuation,especially copper mining,has been the pillar industry in Dongchuan,accompanied by phosphorus mining and iron mining.The copper mining and ore dressing areas are mainly located in Tang Dan,Snow,and Yinmin(Gao et al.2012)(Fig.1).

2.2 Field measurement of heavy metals and sampling

To perform the environmental screening on heavy metals in the study area,12 plots on the fluvial fans along the Xiaojiang River,in a total of 322 farmland,forest land and sediment samples were detected in the field survey using a handheld X-ray fluorescent(XRF)analyzer(Niton XL3t 600+,Thermo Scientific).At each plot,every 50 m set a sample point.A total of 297 surface soil samples were detected in situ(61 from the upstream,136 from the midstream,100 from the downstream)(Fig.2).

Under the guidance of the field measurement of heavy metals on the river fans,six sites were chosen to represent a range of geographical positions on the Xiaojiang River that lie downstream of the main mining zone(Fig.1).In total,54 river water,surface soil,and rice samples were collected at these sites.At each site,three surface soil samples(0—20 cm in soil depth)were collected using a hand corer,and the samples were mixed well to form a composite sample.We focused on surface soils because their heavy metal levels may reflect current environmental conditions(Bai et al.2011).Concurrently,rice samples were collected from the corresponding paddy soil sampling plots.At each site,three replications were combined to form a single composite sample.After collection,these samples were put into polyethylene bags and then immediately transferred to the laboratory for sample processing. Irrigation water samples were collected from the Xiaojiang River.These water samples were stored in pre-acid-washed polyethylene bottles and acidified with concentrated nitric acid to a pH lower than 1；then,the samples were preserved at 4°C in a cooler. The heavy metal concentrations of soils in sequentially extracted forms and rice in the various organs collected from these sites were measured to further understand the effect of copper mining on rice agrosystem.

Fig.1 Study area and location of the sampling stations

2.3 Preparation and analysis of samples

The air-dried soil samples, after removing roots and gravels,were crushed and ground in an agate mortar,and the soils were passed through a 100-mesh sieve and preserved for use.The rice samples were carefully washed with tap water to clear the remnants of the soils and then air-dried and preserved separately as roots,stems,leaves,and grains.All samples were sealed in polyethylene bags and kept at 4°C in a refrigerator before chemical analysis.

Fig.2 Location of field measurement in Dongchuan,Yunnan

Soil pH was measured in a water mixture[soil(mass):water (volume)=1: 5] using an electronic pH meter(Mettler Toledo FE20,Switzerland).Soil organic matter(SOM)was measured by the potassium dichromate oxidation method(Nelson and Sommers 1982).The amount of CaCO3(including other carbonates)in the bulk soil samples was determined using the rapid titration method of Hutchinson and Maclennan(Trivedy and Goel 1984).The cation exchange capacity(CEC)of the mineral soils was calculated as the sum of Ca+Mg+K+Na+Fe+Al+Mn extractable with 1 M NH4-acetate(IGAC 1979).

Heavy metals in the soils were sequentially extracted using the Bureau Community of Reference(BCR),which operationally divided the heavy metals into a weak acid soluble fraction(F1),reducible fraction(F2),oxidizable fraction(F3),and residual fraction(F4)(Rauret et al.1999；Sahuquillo et al.1999；Liu et al.2014).In addition,the bioavailability of metals in the soil samples was measured by the diethylenetriaminepentaacetic acid(DTPA)method(Lindsay and Norvell 1978).

To measure the content of heavy metals in soils and plants,soil samples were digested with an HClO4-HNO3-HF mixture,and plant samples were digested with an HClO4-HNO3mixture in Teflon tubes(Huang et al.2006；Park and Choi 2013).Subsequently,the concentrations of 9 heavy metals(i.e.,Cu Zn,Pb,Cr,Cd,V,Co,Ni,and U)in the water,soil,and rice samples were determined using inductively coupled plasma atomic emission spectrometry(ICP-AES,Perkin Elmer Optima 5300DV,USA).

The accuracies of the heavy metal concentrations were verified with certified values(GBW07401,China National Center for Standard Materials),including water(GB3838-2002),soil(GB15618-2008),and rice(GB2715-2005).

2.4 Pollution assessment methods

The geo-accumulation index(Igeo)and risk assessment code(RAC)were applied to evaluate the contamination of and environmental risks to soil samples from heavy metals.The quality of rice was assessed by the pollution index(Pi)and the Nemerow integrated pollution index (NIPI).Moreover,health risk[hazard quotient(HQ)and hazard index(HI)]was assessed to estimate health risk to the habitats caused by the ingestion of agricultural products.

2.4.1 Geo-accumulation index(Igeo)

The Igeois an effective tool to assess the degree of heavy metal contamination of sediment or soil(Zhan et al.2014；Tian et al. 2017). This method compares the metal concentrations of studied soils to those of the background values,thus providing an important tool for determining the accumulation of the target contaminants(Li et al.2014). The formula proposed by Mu¨ller (1969) is as follows:

where CMi,Biand i represent the total metal concentration in the investigated soil,the background value of the metal element and the measured element,respectively.The Igeovalue for each heavy metal is used to determine its contamination level(Wei and Yang 2010；Li et al.2014).Based on the Igeovalue,the contamination level for a certain heavy metal may be classified as class 1(uncontaminated):Igeo≤0,class 2(uncontaminated to slightly contaminated): 0 ＜Igeo＜1, class 3 (slightly contaminated):1 ＜Igeo＜2,class 4(slightly to heavily contaminated): 2 ＜Igeo＜3, class 5 (highly contaminated):3 ＜Igeo＜4,class 6(highly to severely contaminated):4 ＜Igeo＜5,or class 7:Igeo≥5(Massaquoi et al.2015).

2.4.2 Risk assessment code

The RAC,defined as the percentage of the weak acid soluble fraction,which indicates that the metals are weakly bound to the solid phase,has been used to assess the environmental risk of heavy metal pollution in sediments and soils(Rodr?′guez et al.2009；Nemati et al.2011).Hence,metals pose a greater risk to the aquatic environment due to their greater potential.No risk is present when the RAC value is less than 1%.The environmental risk is presumably low,medium,high and very high when the RAC values are 1—10%,11—30%,30—50%,and ＞50%,respectively.

2.4.3 Pollution indicators

The Pi was used to assess the quality of rice.Following Sundaray et al.(2011),the Pi was defined as:

where Ci is the concentration of a given metal in rice(mg·kg-1),while Ce is the standard value of each heavy metal(mg·kg-1).In addition,to assess the overall pollution status for the rice grain samples,the NIPI(Nemerow and Wolfert 1985)was employed(Shi et al.2014).The NIPI can be derived using Eq.(3):

where avg(Pi)and max(Pi)are the average and maximum values of all Pi,respectively,of the metals being studied.Based on Chai and He(2006),Pi is divided into four levels to indicate the pollution degree,which is classified as follows: unpolluted (Pi ≤1), slightly polluted (1 ＜Pi≤2),moderately polluted(2 ＜Pi ≤3),and highly polluted (Pi ＞3). However, the classification of NIPI is slightly different from the Pi levels and can be classified as safe(NIPI ≤0.7),precaution(0.7 ＜NIPI ≤1.0),slight pollution (1.0 ＜NIPI ≤2.0), moderate pollution(2.0 ＜NIPI ≤3.0), and heavy pollution (NIPI ＞3.0)(Cheng et al.2014).

2.4.4 Health risk assessment

To estimate the intake of heavy metals and adverse health effects on the human body from being exposed to heavy metals through ingesting rice for food,general exposure equations provided by the United States Environmental Protection Agency(USEPA)for health risk assessment were explored(USEPA 2016).The residents who lived in the affected mining area were clustered into two groups:children and adults. The average daily intake (ADI)(mg kg-1day-1)equation was used to calculate the human exposure level.The ADI is defined as:

where C is the heavy metal concentration(mg·kg-1)in rice grain.The average body weight(BW)is assumed to be 60 kg for Chinese adults(Li et al.2013).The average time(AT),as well as exposure duration(ED),were suggested to be 71.4×365 day and 71.4 year(Li et al.2013).Other parameters referred to the USEPA(2016)were as follow:0.42 kg·day-1for ingestion rate(IR)and 350 day·year-1for exposure frequency(EF),respectively.

The health risk of agricultural product intake is typically characterized by the HQ and HI.The HQ is the ratio of the ADI for heavy metals to its reference dose(RfD)(USEPA 1998；Li et al.2014).All HQs can be added to generate an HI to estimate the total potential health risk.Therefore,the HQ and HI can be defined as:

where RfD is the exposure RfD for heavy metals(mg·kg-1·day-1).According to Xiao et al.(2017),when the HQ is lower than 1,there is no adverse effect on human health,whereas adverse health effects may occur when the HQ is greater than 1.According to the USEPA(2007),it is possible that carcinogenic effects may occur when the HI ＞1,while the exposed individual is unlikely to experience obvious adverse health effects when the HI ＜1.

2.4.5 Biological concentration factor

The biological concentration factor(BCF)was a factor indicating heavy metal uptake and bioaccumulation from soil to plant organs(root/shoot)and was calculated as follows:

where heavy metal concentrations in the plant body(Cplant)and soil(Csoil)(Mo et al.2002).

2.5 Statistical analyses

Because most of the data were not normally distributed,non-parametric statistics were applied in the present study.Correlation coefficient analysis was used to estimate the relationships among heavy metal concentrations in the soil,water,and rice and between metal concentrations in the soil and soil properties.The criteria for significance in the procedures were set at p ＜0.05(significant)and p ＜0.01(highly significant).All statistical analyses and manipulations were conducted using the SPSS 20.0 statistical software package and Microsoft Excel 2007.

3 Results

3.1 Heavy metal concentrations in soils and sediments tested in-site by portable XRF analyzer

Across the study area,high levels of Cu,Zn,and V contents were found in most soils.The concentration of heavy metals showed considerable variability in the upstream,midstream,and downstream(Fig.2).The highest Cu(850.00 mg·kg-1),Zn(509.00 mg·kg-1),and Pb(110.00 mg·kg-1)concentrations were observed in the farmlands along the river downstream.The mean concentrations of Cu and V in soils through the whole drainage were significantly higher than its maximum allowance concentration(MACs)set by the Chinese Environmental Protecting Administration (CEPA,1995)for soils in China and soils around other copper mine area,e.g.,China(Cai et al.2015)and Georgia(Avkopashvili et al.2017).

Comparing these values with the contents in forest lands,which were not irrigated with the river water,the Cu concentrations in the long-term irrigated farmlands were much higher(Fig.3).Moreover,the contents of Cu and Zn existed a basic trend that the midstream and downstream from copper mining area were much larger than the upstream.The results could be attributed to the irrigation of the Xiaojiang River to the farmlands.

Fig.3 Heavy metal concentrations in soils and sediments measured by Niton XRF.Error bars indicate the standard deviation.Different capital letters indicate significant differences(p ＜0.05)of the same region between different land use type,while different lower-case letters illustrate significant differences of the same land use type between soils distributed in the upstream,midstream and downstream

It should be noted that soil moisture content had a great impact on the field determination of the heavy metal concentration using the portable XRF instrument(Li 2004).The higher moisture content in soil would result in lowering the detected concentration of the elements(Hu 2016；Turner et al.2017),and the paddy soil moisture capacity reached saturation due to the submerged condition during a field investigation in Mid-May.Thus,the actual heavy metal contamination of soils and the effect of copper mining on heavy metals in a rice agrosystem needed further determination in the laboratory.Nevertheless,the field measurements provided an overview of soil and sediment heavy metals in the Xiaojiang River basin.

3.2 Heavy metal concentrations of water samples

The pH value of all water samples varied narrowly,ranging between 6.45 and 7.18(Table 1)and were lower than the pH values between 7.60 and 8.50 in the Xiaojiang River presented by Huang et al.(2017).As presented in Table 2,the concentrations of dissolved Cu in the water ranged from 23.69 to 91.23 μg·L-1,with an average of 53.82 μg·L-1.The concentration s of Zn in the water averaged 563.61 μg·L-1and ranged from 367.24 to 799.05 μg·L-1.The mean concentrations of dissolved heavy metals in the water samples varied in the order of Zn ＞Pb ＞Cu ＞Ni ＞V ＞Cr ＞Cd ＞U ＞Co(Table 2).The concentrations of Cu,Zn,and Cr in the water were below the maximum limit for surface water quality established by the Chinese National Environmental Quality Standards (SEPA and AQSIQ 2002).Nevertheless,the concentration of Pb was much higher than the maximum concentration permitted by the Chinese National Standard.In comparing these water samples with those from the Panzhihua watershed in the Jinsha River(Zhao 1982),the concentrations of heavy metals were relatively high in the Xiaojiang River.

The variations of the nine trace metal concentrations in water samples were exhibited in Fig.4.Increases in concentrations of Pb,Cd,and Ni in the Xiaojiang River were evident along the mining area and further downstream.In contrast,the concentrations of Zn,Cu,and V in the river water decreased from the middle reaches to downstream,suggesting possible precipitation of the elements from the water into the sediments(Ding et al.2005).

3.3 Heavy metal concentrations in soils

3.3.1 Basic physicochemical characteristics of soils

Fig.4 Heavy metal concentrations in water samples from the Xiaojiang River at the sampling sites

The soil samples exhibited a neutral pH between 6.45 and 7.10 (Table 1). The percent of organic matter in the analyzed soils was low,varying between 1.32%and 1.86%.The contents of carbonates in the soils varied between 13.06%and 20.48%.The soil CEC value varied greatly within a range of 63.00 and 243.60 mmol·kg-1,with an average of 122.90 mmol·kg-1.Reducing substances in the soil mainly included ferrous iron, divalent manganese,active reducing substances and active organic reducing substances,and the concentrations of ferrous iron,divalent manganese, active reducing substances and active organic reducing substances,and total reducing substances were 1.30—5.90,1.00—3.30,7.50—10.20,1.60—8.30,and 17.70—48.20 mmol kg-1,respectively.

3.3.2 Total heavy metal concentration

The heavy metal concentrations in soils detected by ICPAES method were substantially higher than those detected by portable XRF analyzer in situ.This finding was in agreement with previous studies that ICP-AES returned significantly higher concentrations than XRF approaches(Cheng et al.2013；Turner et al.2017).The heavy metal concentrations in the soils were substantially higher than those in the water.Table 2 shows the mean concentrations and the ranges of the considered metals in the soil samples from the 6 sites.Soil metal concentrations occurred in the following order:V ＞Zn ＞Cu ＞Cr ＞Ni ＞Pb ＞Co ＞Cd ＞U,though their concentrations varied depending on the sites.The mean and minimum concentrations of each metal remarkably exceeded the corresponding background values of soils in Yunnan Province and most other soils across the world(Cos?kun et al.2006).According to Grade II of the Environmental Quality Standard for soils issued by the Ministry of Environmental Protection of China,in comparison with the maximum permissible concentrations of chemical elements for agricultural soils,the mean concentrations of Cu,Zn,Pb,Cd,V,and Ni were 6.9,4.2,2.9,29,13.3 and 2.6 times greater,respectively,than their respective maximum allowable concentrations(MACs)for agricultural soils, which indicates that the soils were unsuitable for agricultural production.In comparison with other copper mining areas,the levels of all trace elements in this study were much higher than those found in agricultural soils of the Kilembe copper mine(Abraham and Susan 2017),while the concentrations of Cu and Ni were lower than those of the Jinchuan copper mine in Gansu Province,China(Feng et al.2011).Furthermore,the Cu and Zn concentrations of the surface soil samples at all sites exceeded the MACs for agricultural soils.

The concentrations of the elements in the topsoil,except Cd and U,obviously changed at different sites,and topsoil Pb,Zn,Co,Ni,and V concentrations decreased from the middle reaches to the lower reaches of the Xiaojiang River(Fig.5).Comparatively,the topsoil concentrations of Cu and Cr were highest downstream and lower in the middle reaches of the river.Moreover,samples M2 and M7 had very high concentrations of Pb and Zn,likely because they were closest to the lead—zinc mine,which was located in the middle of the Xiaojiang River(Fig.1).

3.3.3 Heavy metal fractions using the BCR sequential extraction method

Fig.5 Heavy metal concentrations in soils at the sampling sites

In this study, the modified BCR four-step sequential extraction method was used to determine the metal fractions in the contaminated soils.Among the four fractions of metals,the sums of the weak acid-soluble,reducible,and oxidizable fractions are typically considered the mobile and potentially bioavailable fractions for living organisms(Chen et al.2000；Ghayoraneh and Qishlaqi 2017).The predominant chemical forms of Zn,Cr,V,Co,Ni,and U were residual fractions, accounting for 31.36—45.01%,63.04—82.92%, 41.12—66.69%, 35.54—57.06%,45.22—58.89%,and 45.83—69.17%of their total concentrations,respectively.These results were similar to the results of Ferna~(2017),who conducted his research in a copper mining area in Touro(Galicia,NW Spain),where the predominant forms of Zn,Cr,and Ni were residual fractions.Heavy metals in residuals are unlikely to be discomposed under normal environmental conditions and might be considered stable heavy metals.In terms of geographical distribution,residual fractions of these heavy metals decreased along the Xiaojiang River.

Cd was mainly present in the weak acid soluble fraction,accounting for up to an average of 19.6%of the total concentration in the studied soil(Fig.6),which implies the relatively high mobility and bioavailability in our studied soil.At the same time,Pb primarily existed in reducible and oxidizable fractions,suggesting that the major mechanism controlling the fate and transport of Pb was trapped by Fe—Mn oxides(Kassir et al.2012).The chemical distribution of Cu in the topsoil depended on the sites.In the middle reaches,Cu speciation primarily existed in the residual fraction,while the reducible fraction was dominant in the topsoil downstream. These results are partially consistent with the data gathered in a lead—zinc mine area located in Chenzhou, Hunan Province, China (Huang 2014),that demonstrated that the predominant form of Cd was the weak acid soluble fraction,and the predominant forms of Cu were the residual fraction and oxidizable fraction,possibly due to the existence of the lead—zinc mine in the area.

3.3.4 Heavy metal fractions using the DTPA method

The bioavailable amounts of heavy metals determined by the DTPA extraction procedure primarily depend on the type of metal involved and the conditions of the sampling sites.Different heavy metals exhibit various metal availabilities.The Cu DTPA-extractable fraction varied from 12.50 to 560.00 mg·kg-1,while the Cr value ranged from 0.70 to 1.50 mg·kg-1(Table 3)data based on the results from the soil samples of this area,Cu is the most abundant metal in a bioavailable form (DTPA-extractable value divided by total concentration),followed by Zn,Pb,Cd,and Co(Table 3).

3.4 Heavy metal concentrations in rice

3.4.1 Heavy metal concentrations in rice grains

Plants cultivated in polluted soils are affected by trace metal pollution(Chary et al.2008；Zhao et al.2010；Khillare et al.2012；Minkina et al.2012).The uptake of heavy metals varies with different crop species(Pepper et al.1983).As shown in Table 2, the Cu concentrations ranged from 43.00 to 547.00 mg·kg-1in the rice grains,with an average of 238.00 mg·kg-1.The Zn concentrations ranged from 179.00 to 2322.00 mg·kg-1in the rice grains, with an average of 1518.00 mg·kg-1.By comparing the results with other mining areas,the mean concentrations of Cu,Pb,and Zn in rice vastly exceeded the values(238.00,373.00,and 1257.00 mg·kg-1,respectively)obtained from four representative abandoned metal mining areas in South Korea(Ji et al.2013).Additionally,the mean concentrations of Cu and Zn were much higher in the rice grains than those of other plants obtained from a copper mine area in Cyprus(Christou et al.2017),probably due to the high concentrations of heavy metals in the soils we studied.

Fig.6 The percentage of heavy metal fractions in the soils at the sampling sites in Dongchuan,Yunnan

To determine the extent of contamination in crops grown in soils affected by mining,the concentrations in the crop samples were compared with the corresponding maximum allowable level(MAL)of metals in vegetables and cereals in China(State Environmental Protection Administration of China(SEPAC)2005).According to the State Environmental Protection Agency of China(SEPAC)(2005),the MALs of Cu,Pb,and Cd for vegetables and cereals are 0.20,10.00 and 0.20 mg·kg-1,respectively,on a fresh weight basis.The average concentrations of Cu,Zn,Pb,Cr,Cd,and Ni in the rice grains were 24,25,1865,88,22,and 40 times higher,respectively than their MALs.Additionally,the concentrations of these heavy metals at all the sites exceeded the corresponding MALs, highlighting the potentially hazardous effects of consuming products that grow on contaminated soil.

3.4.2 Heavy metal concentrations in roots,leaves,and stems

The change in heavy metal concentrations in different parts of the rice sampled along the Xiaojiang River was in the order of roots ＞grains ＞leaves ＞stems (Fig.7),which is different from the order of roots ＞stems ＞grains represented in the Dabaoshan mine area in Guangdong Province, China (Liao et al. 2016). The higher concentrations of heavy metals in the roots,with average concentrations of Cd, Cr, and Pb at 25.35,749.17,and 1066.74 mg·kg-1,respectively,than those in the other parts of the rice indicated that roots have strong absorption ability and produce an obvious barrier effect(Meng et al.2016).Specifically,zinc was uniform in its distribution downstream. The concentrations of these heavy metals,except for Pb,in the rice grains were higher downstream than mid-stream.

Fig.6 continued

4 Discussion

4.1 Effect of copper mining on heavy metals in rice agrosystems

4.1.1 Water quality

The given data indicated that the concentrations of heavy metals in Xiaojiang River water remained at relatively low levels,except for Pb,which far exceeded the maximal concentration permitted by the National Standard of China(SEPA and AQSIQ 2002).In comparison with the values in other parts of the Jinsha River Basin and different copper mining areas,the concentration of heavy metals was much higher in this study(Zhao 1982；Abraham and Susan 2017).The 2016 heavy metal concentrations in the water of the Xiaojiang River were greatly lower than those in this study(Huang et al.2017)；this difference can be attributedto the closure of most mines in Dongchuan since the Milk River event in 2013(Xie 2013).Nevertheless,the high metal concentrations (334.53—344.53 mg kg-1for Zn,218.69—220.63 mg kg-1for Ni,etc.)in the sediments from the Xiaojiang River indicate the profound effect of mining on the fluvial system(Ding et al.2005；Zhang 2017).

Table 3 DTPA-extractable values for different soil samples

Fig.7 Heavy metal concentrations in root and shoot of cultivated plants sampled from various sites

4.1.2 Soil

Many studies, e.g., those in China (Liu and Probst 2005a,b；Cai et al.2015),Romania(Harmanescu et al.2011),Spain(A′lvarez-Ayuso et al.2012),Nigeria(Obiora et al.2016),and Georgia(Avkopashvili et al.2017),have noted that heavy metals in agricultural soils near metal mines accumulate to fairly high concentrations.Consistent with these previous studies,the present study found that all heavy metal concentrations in riparian soils of the Xiaojiang River were quite high and unsuitable for crop production.The fluvial soils of the studied area,derived mainly from carbonate rocks and other types of lithology(Fig.1),initially should have had relatively low heavy metal concentrations (e.g., 14.60 mg·kg-1for Pb and 0.6 mg·kg-1for Cd). Therefore, the high soil concentrations of these metals were primarily affected by using river water for land irrigation in the area(Yang et al.2008).To investigate the relationship between heavy metal concentrations in fluvial soil and water in the Xiaojiang River reaches,Pearson's correlations between metal concentrations in the soil and water and other properties of the soils were analyzed(Table 4).The nearly identical and significant correlation between soil and river water pH values illustrated that soil pH was controlled by long-term irrigation with river water(Table 1),consistent with the reports by Ma et al.(2015)and Andrews et al.(2016).However,it seems that high concentrations of heavy metals in the soils were derived from the deposition of sediments mixed with mining tails and irrigation of with water from the Xiaojiang River.The average heavy metal concentrations in the sediments of the same research area were 70.17 mg·kg-1for Cu, 339.53 mg·kg-1for Zn and 17.07 mg·kg-1for Co(Zhang 2017).In addition,riverbed and suspended sediments in the Jinsha River(the mainstream of the Xiaojiang River)were enriched with heavy metals；for example,the enrichment factor of Cd in the sediments was as high as 13,000(Zhao 1982).The accumulation of sediments carried by river water can result in an increase in heavy metal concentrations in soils due to the dissolution of carbonate minerals in soils under an acidic to neutral pH(Yu and Lai 2006).

Fig.7 continued

4.1.3 Rice grains

Crops grown in soils affected by mining may take up and accumulate heavy metals at an excess rate(Zhuang et al.2009；Christou et al.2017).In our study,the accumulation of Cu,Pb,and Cd in the soils due to long-term water irrigation increased the concentrations of Cu,Pb,and Cd in the rice grains in the Xiaojiang River Basin. It is inevitable that long-term mine wastewater irrigation leads to heavy metal pollution in soil-crop systems.However,the heavy metal concentrations in the rice grains were not proportional to those in the water when these results were compared to the results in Ma et al.(2015),who investigated heavy metal uptake by wheat in a coal-mining field.The accumulation of heavy metals in rice grains involves the uptake of heavy metals by the roots from the soil,the subsequent transport of heavy metals from the roots to the shoots via the xylem and the translocation of heavy metals from the xylem to the phloem(Clemens et al.2002),which depends on the heavy metal concentration in the soils and soil properties(Rafiq et al.2014；Xu et al.2015；Yang et al.2015).The effects of soil properties on heavy metal uptake might be limited in this study since our findings revealed that the major soil properties were not highly variable(pH:6.45—7.10 and organic matter:1.32—1.86 g·kg-1).The BCF values between the soil and roots for different heavy metals for rice are shown in Fig.8 and were much higher than 1.00,with Pb at 4.51,Zn at 3.04,Cd at 1.65,Cr at 3.10,U at 1.11,and Cu at 2.12.However,most BCFs among the stems and leaves were lower than 1.00,indicating that root accumulation and migration was the main pathway for heavy metal transfer from the soil into paddy rice.The elevated heavy metal concentrations in the rice grains were comparable to those in previous studies(Ji et al.2013；Liao et al.2016).With high metal concentrations found in contaminated paddy soils,high heavy metal concentrations could occur in rice grains.Thus,the decrease in crop quality may have resulted from irrigation with river waterand soils with high concentrations of heavy metals that were impacted by mining.

Table 4 Correlation matrix between heavy metal concentrations and soil properties

4.2 Heavy metal contamination assessment in the agrosystem

4.2.1 Heavy metal pollution and environmental risk evaluation in soils

Heavy metal contamination and environmental risk of the agricultural soils were assessed using the Igeoand RAC,respectively.The Igeovalues at each sampling site for individual elements are shown in Table 5.The Igeovalues of different elements varied with the sampling site and were related to the distribution of heavy metal concentrations in the topsoil.The pollution severity of the heavy metals was in the order of Cd ＞Cu ＞V ＞Zn ＞Ni ＞Pb ＞Cr ＞U ＞Co.The Igeovalues ranged from 2.16 to 4.16 for Cu and 2.47 to 3.22 for Zn.The mean Igeovalues of all the samples for Cu,Zn,and Pb were 3.13,2.93,and 1.94,respectively,indicating that the elevated heavy metal concentrations in the soils were associated with mining activities.The results also revealed that the soils in the studied areas were severely contaminated(class 7)by Cd；highly contaminated(class 5)by Cu；slightly to heavily contaminated(class 4)by Zn and V；slightly contaminated(class 3)by Pb,Cr,Ni,and U；uncontaminated to slightly contaminated(class 2)by Co；and unpolluted(class 1).The average Igeovalues of Cu and Zn in the studied soils were obviously higher than those in the paddy soils near the Barapukuria coal mine in northwest Bangladesh(Halim et al.2015).In addition,the average Igeovalues of Cd,Zn,and Pb also exceeded the values in the Dabaoshan copper mine(Liao et al.2016).

Fig.8 Migration of heavy metals from the soil to different parts of the rice

The percentage of bioavailable heavy metals is critical for assessing environmental risk and RAC；it represents the proportions of bioavailable fractions of heavy metals in the soil and can indicate environmental risk(Liu et al.2013).The ecological risk(RAC)levels of these heavy metals were ranked in the following order:Cd ＞Zn ＞Co ＞Cu ＞Ni ＞Pb ＞Cr ＞U ＞V(Table 5).Generally,a low risk for Cu,Pb,Cr,V,Co,Ni,and U occurred in soils with RAC values less than 10%,which indicated insignificant metal mobility for these elements(Tang et al.2017),whereas a medium risk was indicated for Zn and Cd.The RAC results indicated that high proportions of Cd and Zn can transfer from agricultural soils to crops(Liu et al.2013).Moreover,higher RAC values of Cu,Pb,Zn,Cd,Cr,Co,Ni,and U were found in the downstream of the Xiaojiang River than those in the middle reaches.The variation in RAC values for heavy metals is consistent with the heavy metal speciation concentrations shown in Sect.3.3.3.

Thus,the agricultural soils in this study were mainly affected by Zn and Cd and had a moderate level of pollution based on the two evaluation methods.As shown in Table 5,there were substantial differences between the results of the two evaluation methods,except for Co,which expressed similar slight contamination in both methods.The evaluation of the Igeoindicated that most of the soils were moderately to severely polluted with heavy metals.However,the risk assessment showed that the soils were less polluted by heavy metals.For example,the soils were moderately contaminated with Pb,Cr,Ni,and U based on the Igeo,while the soils were slightly contaminated by these heavy metals based on the RAC.These heavy metals originated from the minerals in the mine and were difficult to release,reducing the extent of their ecological risk；these factors likely caused the inconsistencies in the results(Guo et al.2013).

4.2.2 Heavy metal contamination in rice grains

To assess the intensity of heavy metal contamination in the rice,the Pi and the NIPI were calculated.On the basis of the quality standard for food in China,the Pi and NIPI values of heavy metals may reveal the extent of rice pollution(SEPAC 2005).Overall,the order of the mean Pi values was Pb (1863)＞Cr (101)＞Cd (58)＞Ni(46)＞Zn(24)＞Cu(20)(Table 6).The results showed that the soils were heavily polluted with these heavy metals(Pi ＞3).When comparing the contamination degree of soils or crops at different sampling sites,the Pi values of Cu,Zn,Cr,Cd,and Ni were much higher downstream relative to those values in the middle reaches,while that of Pb was highest in the middle reaches.The spatial distribution of these elements is similar to that indicated by the RAC,thus relating to the bioavailability of heavy metals utilized by plants.

Table 6 The pollution index(Pi)and the Nemerow integrated pollution index(NIPI),hazard quotient(HQ)and hazard index(HI)

The NIPI values of the heavy metals in the rice in the Dongchuan copper mine area can be arranged in descending order: M7 (2847)＞D6 (1990)＞D4 (965)＞M3(942)＞D5(852)＞M2(423).The NIPI values of the rice grains at all sampling sites exceeded 3,also indicating severe pollution with heavy metals(Table 6).The maximum NIPI value reached 2786,which was detected at Site M7,with Pi values of 3980 for Pb,114 for Cr,45 for Cd,35 for Ni,18.32 for Zn and 14.4 for Cu.In general,heavy metal pollution was severe in rice grains and needs to be strictly controlled.

4.3 Health risks for local residents

These metals were not only toxic to the crops but also harmful to animals and humans(Alloway et al.1990).Human exposure to heavy metal pollution derived from the Dongchuan mines might occur directly through ingestion,dermal contact and inhalation of soil or indirectly through consumption of locally grown crop/vegetables. The chemical elements such as Cd,Cr,and Pb in this study had carcinogenic effects on human populations(WHO 2011).Health risks to residents in the study area through the consumption of rice grains were assessed by estimating the HQ and HI.An HQ value greater than 1 indicates that potential health risk may be present.As seen from Table 6,the HQs of Cu,Zn,Pb,Cr,Cd,and Ni ranged from 7.17 to 63.47,4.01 to 46.15,198.16 to 1335.74,0.11 to 1.65,58.87 to 119.55,0.73 to 14.38,respectively.The HQ values of the heavy metals except Cr were all beyond the safe threshold (HQ=1), suggesting that direct impacts of heavy metal contamination may be significant for local resident health(Cai et al.2015).In contrast to previous studies,the health risk for adults in the study area was much higher than that near Pb—Zn mining areas in Enyigba,southeastern Nigeria(Obiora et al.2016),and the Tonglushan mine in Hubei,China(Cai et al.2015).

Considering all the metals, the highest HQ value,1335.74,was observed for Pb in M7.The mean HQs for heavy metals decreased in the following order: Pb ＞Cd ＞Cu ＞Ni ＞Zn ＞Cr.These results indicate that Pb was the major metal contributing to the potential health risk,followed by Cd and Cu.Previous studies have noted that Cd and Pb tend to accumulate in rice grain,which is the most important source of dietary Cd and Pb in China(Huang et al.2018).The HQs of Cr via the consumption of rice grains in the study area were generally less than 1,suggesting that the local residents are exposed to potential health risk from dietary Cr.In addition,the HI values of rice grains varied from 270.57 to 1485.94 at the different sampling sites,with a mean value of 777.09,which is extremely higher than the safe level (HI ＞10). Thus,exposure to two or more contaminants may trigger additive and/or interactive effects,which indicates an increased risk of adverse health effects(Yi et al.2011；Ji et al.2013).

In this study,the Igeo,RAC,Pi and HQ values of each tested metal were used to assess the heavy metal pollution in the rice agrosystem(Table 7).The Igeovalues of Cu,Cd,Pb,Zn,Ni,Cr,U,and V in the soils demonstrated moderate to severe contamination.In addition,rice grain contamination levels of Cu,Cd,Pb,Zn,Ni,and Cr were severe,and further consumption of rice grains would result in substantial potential hazards from Cu,Cd,Pb,Zn and Ni to the local residents. However, a substantial difference appeared between the RAC and the other indicators.The RAC values for Cd and Zn in the soils were moderate,whereas the contamination levels of Cu,Pb,Cr,Co,Ni,U,and V were slightly contaminated to uncontaminated,even though the concentration of weak acid soluble fractions ofmetals was high.This result could relate to the fact that the RAC indicator is based on the proportion of the weak acid soluble fraction to the total heavy metal fractions rather than on the concentration.In addition,the Pi of NIPI and HI suggested that all sampling sites were severely contaminated by heavy metals, especially in the middle reaches,which indicated an intense effect of copper mining on the rice agrosystem.

Table 7 The pollution level of the rice in the agrosystem based on Igeo,RAC,Pi,HQ,NIPI and HI

5 Conclusions

These findings unequivocally showed that copper mining had a great impact on heavy metals in a rice agrosystem near the Dongchuan copper mining area,Yunnan,southwest China.Long-term wastewater irrigation,especially the deposition of sediments, lowered the soil pH and induced the accumulation of heavy metals in the fluvial soils along the Xiaojiang River.The Cu,Zn,Pb,Cr,Cd,and Ni concentrations in the soils far exceeded the MACs for agricultural soil,and the accumulated heavy metal concentrations in the rice grains partially or completely surpassed the MALs set by China.The heavy to severe pollution of the agricultural soil with Cd,Cu,Zn,and V was evident based on the pollution indicator(Igeo)values.Nevertheless,only soil Cd and Zn concentrations were of modest risk based on the RAC values.The difference in the pollution evaluation methods using these two indicators probably related to the reduction in metal bioavailability due to the immobilization of heavy metals in carbonate minerals,resulting in a decrease in RAC values.The high Pi value showed that rice grains are essentially contaminated with Pb,Cr,Cd,Ni,Zn,and Cu,of which Pb is the major contaminant.The HQ values,which were used to evaluate the carcinogenic risk of the studied soils and crops on human health in terms of heavy metals,were higher than 1 except for the HQ of Cr, indicating a severe potential health risk from Cu,Pb,Zn,Cd,and Ni to local residents.Among the all sampling sites analyzed,the sites in the downstream of the Xiaojiang River were generally more contaminated than those in the middle reaches.Overall,the results suggest that rice agrosystems around mining areas were highly polluted,mainly with Cd,Zn and Cu；long-term heavy metal pollution and related health effects should be of continuing concern in mining-affected areas due to the persistence and high toxicity of heavy metals. In summary, long-term mining activities, like copper mining in Dongchuan,have negatively impacted the surrounding soils,crops,rivers and streams,and increased health risk for local residents.Effective measures should be taken to protect the environment and ecosystem against the mining pollution and remediate contamination in the mining zones and adjacent regions.

AcknowledgementsThis work was funded by National Key Research and Development Program of China (Grant No.2017YFC0504902)and National Key Technology R&D Program of China(Grant No.2012BAC06B02).

Compliance with ethical standards

Conflicts of interestThe authors declare that they have no conflict of interest.