- Email: [email protected]
Distribution and availability of cadmium in profile and aggregates of a paddy soil with 30-year fertilization and its impact on Cd accumulation in rice plant
Environmental Pollution 239 (2018) 198e204
Contents lists available at ScienceDirect
Environmental Pollution journal homepage: www.elsevier.com/locate/envpol
Distribution and availability of cadmium in profile and aggregates of a paddy soil with 30-year fertilization and its impact on Cd accumulation in rice plant* Zhong-Xiu Rao a, b, Dao-You Huang a, Jin-Shui Wu a, Qi-Hong Zhu a, *, Han-Hua Zhu a, Chao Xu a, Jie Xiong a, b, Hui Wang a, Ming-Meng Duan a, b a
Key Laboratory for Agro-ecological Processes in Subtropical Region, Institute of Subtropical Agriculture, Chinese Academy of Sciences, Changsha, Hunan, 410125, China University of Chinese Academy of Sciences, Beijing, 100049, China
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 17 November 2017 Received in revised form 5 April 2018 Accepted 5 April 2018
The research was conducted to investigate the accumulation, distribution and availability of Cd in paddy soil and their relation to Cd in rice plants under 30-year fertilization regimes. Six treatments were involved in the study: control without fertilization (CK), chemical fertilizer (NPK), high nitrogen chemical fertilizer (HN), rice straw incorporation (ST), low and high dosage of manure fertilizer (LM and HM). Total and DTPA extractable concentration of Cd (T-Cd and DTPA-Cd) in bulk soils (20 cm topsoil), profiles (0 e60 cm) and aggregates (>2, 1e2, 0.5e1, 0.25e0.5, 0.053e0.25 and < 0.053 mm) were investigated. The Cd concentration in relevant rice plant (roots, stems, leaves, husks and grains) were also analyzed. Manure fertilizers caused T-Cd accumulation in bulk soil with a significant increase of 36.2% in LM and 81.2% in HM. Similar impacts of manure fertilizers were observed in DTPA-Cd in the bulk soil. Further, the HM generated a further accumulation in deeper soil layers, presenting a remarkable increase of T-Cd (28.3%e225%) in 10e40 cm and DTPA-Cd (116%e158%) in 10e30 cm profiles. Moreover, the continuous application of manure fertilizers enhanced the availability of Cd in all aggregate size classes with an increase of 17.3%e87.8% in DTPA-Cd. Organic fertilizers (LM, HM and ST) heightened the content of Cd (38.0%e152%) in all parts of rice plant. The accumulation of Cd in rice plants was directly affected by fertilization regimes and Cd availability in the 10e20 cm soil layers and 0.25e0.5 mm aggregates. In conclusion, long-term application of manures resulted in increasing availability of Cd in aggregates and in topsoil and subsoil layers, which accordingly enhanced the accumulation of Cd in rice plants. © 2018 Elsevier Ltd. All rights reserved.
Keywords: Paddy soil Aggregates Profile Rice Cadmium (Cd) Accumulation
1. Introduction Cadmium (Cd) is a hazardous heavy metal found in the environment. Anthropogenic activities have resulted in the release of a substantial amount of Cd into cropland and especially in paddy regions (Grant and Sheppard, 2008). Considering the high transfer ability of Cd in soil-rice system, the contamination risk of Cd in rice grain has been found all over the world, e.g. China and Southeast Asia (Zhao et al., 2015; Tang et al., 2016). Controlling the accumulation of Cd in rice grains have been a worldwide issue because of
*
This paper has been recommended for acceptance by B. Nowack. * Corresponding author. E-mail address: [email protected] (Q.-H. Zhu).
https://doi.org/10.1016/j.envpol.2018.04.024 0269-7491/© 2018 Elsevier Ltd. All rights reserved.
the bioaccumulation of Cd in crops and its potential risks to human body through the food chain (Xu et al., 2015; Chen et al., 2017). Previous studies indicated that Cd-bearing fertilizers, especially manures, may be one of the important sources for Cd entering into soil (Zhang et al., 2011; Cheng et al., 2013; Zhao et al., 2014). For example, livestock manures were observed to account for 55% of the total Cd inputs in agricultural soils of China (Luo et al., 2009). The effect of fertilization on Cd in soil has been related not only to the concentration and availability of Cd in bulk soils but also to the accumulation and distribution of Cd in soil profiles and aggregates (Acosta et al., 2011; Fan et al., 2012). Application of pig manures has been reported to significantly increase the Cd concentration in soils (Wu et al., 2012; Xu et al., 2015). It was also observed that Cd in subsoil layers increased significantly after continuous application of pig manures for 10 years (Xu et al., 2015). Furthermore, the
Z.-X. Rao et al. / Environmental Pollution 239 (2018) 198e204
availability of Cd may be strongly affected by soil aggregate size distribution because aggregates are the basic unit of soil construction and important place for chemical and biological reaction activities (Ilg et al., 2004; Fedotov et al., 2008). Increasing risk of Cd uptake by rice was also found in paddy soils fertilized with pig manure and green manure (Wang and Zhou, 2017). Long-term fertilization is likely to result in much more serious accumulation risks of heavy metals in soils and crops (Hejcman et al., 2008; Zhao et al., 2015; Zhou et al., 2015). It is crucial to determine the distribution of Cd, and its availability, at the profile and aggregate scales, as affected by long-term fertilization regimes. Therefore, the distribution and accumulation characteristics and availability of Cd in bulk paddy soil, profiles, aggregates and rice plants were analyzed using a 30-year fertilization experiment. The objective was to elucidate the effects of long-term fertilization on Cd availability in soil profiles and aggregates and relate their impacts on Cd accumulation in rice plant. 2. Materials and methods 2.1. Site and soil description The long-term fertilization experiment was initiated in 1986 in Hunan Province, China. The site has a humid subtropical monsoon climate, an annual average temperature of 16.8 C and a mean precipitation of 1358 mm. The tested soil was derived from modern river alluvial deposits and classified as alluvial sand soil. The pH of the tested soil is 5.6 on average, and it contains an average quality of soil organic carbon (SOC) 21.1 g kg1, total nitrogen (TN) 2.17 g kg1, total phosphorus (TP) 1.08 g kg1 and total potassium (TK) 22.3 g kg1. 2.2. Fertilization designs The fertilization regimes had been applied and lasted for 30 years up to 2015, in which year the soil samples were collected. The fertilization treatments were: no fertilizer (CK), chemical fertilizer (NPK), high nitrogen chemical fertilizer (HN), rice straw incorporation (ST), low dosage of manure fertilizer (LM) and high dosage of manure fertilizer (HM). The dosage of chemical fertilizer (NPK) was determined according to the N, P and K supplying capacity of the soil and the absorption ability of plants in the supervised field. The source of N and K were urea fertilizer (46%) and potassium chloride fertilizer (60%), respectively. The P was applied in the form of calcium-magnesia phosphate fertilizer (12%). Thus, some other elements such as Ca, Mg and Si were also imported into soil coupled with P. High nitrogen chemical fertilizer (HN) involved further application of N, additional to the NPK treatment. The manure fertilizer used in the study was mainly pig and fowl manure, and the proportions of N in the low and high dosages of manure fertilizer were 30% and 60% of total N applied to the soil, respectively. The amount of ST was 3000 kg ha1. The LM, HM, and ST treatments were supplemented by chemical fertilizers. The details of fertilizer application are shown in Table 1. 2.3. Sampling The bulk soil samples were collected from the 0e20 cm surface soil layer. To ensure the representation and uniformity of soil samples, a composite soil sample was obtained by thoroughly mixing 5 subsamples at each sampling site. Bulk soil samples were pre-treated by air drying and then ground to pass through 1 mm and 0.149 mm sieves for further analysis. Profile samples were collected in the soil layers from depths of 0e60 cm and were divided into 6 layers with 10 cm per layer. A
199
Table 1 Field fertilization amount of tested soils (by purity nutrient content, kg/ha/a) (Di et al., 2014). Treatments
Chemical fertilizer
Straw/Manure
Total amount
N
P
K
N
P
K
N
P
K
NPK HN ST LM HM
457 757 378 313 164
44 44 36 30 36
160 160 1 61 0
0 0 81 147 293
0 0 12 56 111
0 0 167 106 211
457 757 459 460 457
44 44 48 86 147
160 160 168 167 211
*Field fertilization amount (purity content) represented the average values between 2004 and 2011.
stainless steel earth drill with a scale line mark was used for sample collection. The profile samples were air dried and sieved to <0.149 mm for further analysis. Water-stable aggregates were derived from 0 to 20 cm topsoil and separated into six size fractions by wet sieving (Elliott, 1986) as, >2, 1e2, 0.5e1, 0.25e0.5, 0.053e0.25 and < 0.053 mm. Five sequential size sieves were used and placed in descending order from top to bottom. Before fractionation, the soil samples were submersed in distilled water for 10 min and then sieved vertically in distilled water by machine at a vertical reciprocating frequency of 35 r min1 over a period of 15 min. After separation, the samples in the sieve were collected into aluminum specimen boxes, and the samples <0.053 mm in the water were collected by settlement and centrifugation. All aggregate samples were then dried at 60e70 C in an oven at constant temperature. The plant samples were also collected together with the soil samples. The rice plants were collected in their entirety. After deactivation of enzymes at 105 C, the rice plants were then divided into roots, stems, leaves, husks and grains. All the plant sample components were dried in oven at 60e70 C and then ground into powder with a stainless steel pulverizer for further analysis. 2.4. Physical and chemical analysis The soil physical and chemical properties (SOC, pH, and total N/ P/K) were analyzed according to Lu (2000). The total Cd in the bulk soil, water-stable aggregates and soil profiles was solubilized by digestion of 0.5 g in 10 mL aqua regia (HCl: HNO3 ¼ 3:1, V/V; AR) and 2 mL HClO4 (AR). The available Cd was extracted with a DTPATEA-CaCl2 solution by shaking for 2 h at 180 r min1. This DTPA method is widely used and regarded as national standard protocols to investigate soil extractable Cd in China (Zhu et al., 2016). The Cd in rice plants was solubilized by digestion of 1.0 g in 10 mL HNO3 (AR) and 2 mL HClO4 (AR). All detected soil and plant samples were digested or extracted in two duplicate. Distilled deionized water was used for blank correction and the referenced soil (GBW07405) was used as certified reference material. Both of the blank and reference material were digested in the same pattern of soil samples. All digestion and extraction solutions were finally analyzed via inductively coupled plasma optical emission spectrometry (ICPOES, Varian, US), with a limit of 0.003 mg mL1 for Cd. A stock solution of Cd (1000 mg mL1, Merck) was used as an external standard by diluting it into 1 mg mL1. Moreover, all data were blankcorrected and the standard curve was checked every ten samples during the detection procedure. 2.5. Statistical analysis All statistical analyses were performed via Excel 2010 and SPSS
200
Z.-X. Rao et al. / Environmental Pollution 239 (2018) 198e204
19.0 statistical software (SPSS Inc., Chicago, IL, US). Statistically significant differences between the means were tested by an ANOVA within a general linear model (GLM). The least significant difference (LSD) was used to assess the significance at the 95% confidence level. Structural equation modeling (SEM) was performed via Amos 24.0 software. The structural equation modeling (SEM) is a suitable method to provide insight into the relationships of physical, chemical and biological variables in ecological research (Arhonditsis et al., 2006; Grace et al., 2010). It focuses on understanding direct and indirect pathways, to extract scientific understanding about systems (Grace et al., 2010). In the present study, the SEM framework was applied to investigate the direct and indirect effects of Cd bioavailability in soil profiles and aggregates on Cd in rice plants. All data were standardized to achieve linearity and normality. A maximum likelihood method was used to estimate the parameters. The goodnessof-fit of the model was evaluated by a Satarra-Bentler scaled chisquare statistic. The model was rejected if the p-value of the c2 test was below the significance level 0.05. Several variables in the final structure model were excluded because their paths were nonsignificant and inconsistent with the c2 and p-value of the model. The model can be re-specified according to modification indices. A given path can then be added if the modification index is above 4 and scientifically supported (Beaumelle et al., 2016). Several criteria, such as the goodness-of-fit index (GFI), comparative fit index (CFI) and root mean square error of approximation (RMSEA), were used for evaluation of model fit (Hu et al., 2014; Beaumelle et al., 2016). Good model fits were reflected by RMSEA values below 0.05 and GFI values above 0.90. 3. Results 3.1. Accumulation and availability of Cd in bulk soil The total Cd (T-Cd) and DTPA-extracted Cd (DTPA-Cd) in the bulk soil ranged from 0.33 to 0.72 mg kg1 and 0.17e0.36 mg kg1, respectively (Fig. 1). The T-Cd in the control (CK) was 0.40 mg kg1, which was higher than that of the Chinese soil environmental quality standards (Class Ⅱ, 0.3 mg kg1). Compared with CK, the TCd decreased significantly (p < 0.05) by 16.5% in NPK, whereas no significant differences were observed in HN and ST. However, the TCd in HM and LM increased significantly (p < 0.01) by 81.2% and 36.2% compared with that in CK, respectively, and by 116% and 63.1% compared with that in NPK, respectively (Fig. 1). The DTPA-Cd in CK was 0.21 mg kg1, with a proportion of 53.0% of T-Cd. Compared with CK, the DTPA-Cd in HM and LM increased
significantly (p < 0.01) by 73.6% and 28.5%. However, the DTPA-Cd in NPK was significantly lower (18.2%) than that in CK. No significant differences were observed between HN and ST treatments. Compared with the DTPA-Cd in NPK, those in the HM, LM and ST treatments were significantly increased (p < 0.01) by 112%, 57.1% and 24.3%, respectively (Fig. 1). The proportion of DTPA-Cd in T-Cd of all fertilizer treatments was 50.0%e59.5%, with the highest proportion in the ST. 3.2. Distribution and availability of Cd in the soil profiles The T-Cd in the soil profiles of CK was relatively high in the 0e20 cm surface layer and then sharply declined in the profiles below 30 cm, slipped to a range of 0.14e0.19 mg kg1. Compared with CK, the HM had the highest concentration of T-Cd, which showed a significant increase (p < 0.01) of 28.3%e225% in 0e40 cm soil profiles (Fig. 2). In addition, the T-Cd in HM of 0e40 cm profiles was significantly (p < 0.01) 45.7%e395% precede those in NPK, HN and ST (Fig. 2). The T-Cd in 0e10 cm soil layer of LM was significantly higher (p < 0.01) by 23.1% than that in CK (Fig. 2). The DTPA-Cd in CK declined as the depth of the soil layers increased, with little change observed in 30e60 cm soil profiles (Fig. 2). Similarly, the DTPA-Cd in the other fertilized treatments had the same pattern as that of the CK, and the concentration tended to be uniform in 40e60 cm soil profiles. In addition, no significant differences in 20e60 cm profiles were observed for CK, NPK, HN, ST and LM. However, the DTPA-Cd in 0e10 cm profile of the LM and 0e30 cm profiles of the HM increased significantly (p < 0.01) by 48.7% and by 69.3%e157%, respectively, compared with that in the CK (Fig. 2). 3.3. Distribution and availability of Cd in soil aggregates The T-Cd in the aggregates of the CK ranged from 0.27 to 0.80 mg kg1, with the highest concentration in the >2 mm aggregates and the lowest concentration in the <0.053 mm aggregates. Compared to CK, the T-Cd in the aggregates of LM significantly increased by 4.35%e48.6% (p < 0.01) in 0.5e1 mm and 0.053e0.25 mm aggregates. In addition, the T-Cd in the aggregates (except for 0.5e1 mm) of HM (p < 0.01) increased significantly by 13.9%e70.0% (Fig. 3). The T-Cd in >2 mm and 0.5e1 mm aggregates of ST, HN and NPK showed a significant decrease of 12.3%e60.6% compared with that in the CK, whereas an increase (significantly) of 19.6%e77.6% was observed in the <0.053 mm aggregates (Fig. 3). When compared with the T-Cd in NPK and HN, those in the >2 mm aggregates in ST, LM and HM increased gradually and significantly by 14.9%e285% (Fig. 3). Similarly, the T-Cd in the 1e2 mm aggregates increased by 15.8%e41.9% in ST, LM, and LM compared with those in NPK and HN (Fig. 3). The DTPA-Cd content ranged from 0.11 to 0.26 mg kg1 and was relatively high in the 0.25e0.5 mm aggregates followed by the 0.5e1 mm aggregates in the CK. Compared with the DTPA-Cd in the CK, those in LM and HM increased significantly (p < 0.01) by 17.3%e 87.7% and 41.1%e75.9% in all aggregates, respectively (Fig. 3). However, the DTPA-Cd in ST only showed a significant decrease of 21.0% in >2 mm aggregates. Similarly, significant results were found in the 0.25e0.5 mm and 0.053e0.25 mm aggregates of NPK and >2 mm and 0.5e1 mm aggregates of HN (Fig. 3). Compared to the inorganic treatments, DTPA-Cd in the LM and HM increased by 8.97%e153% and 34.5%e124% (p < 0.01) in aggregates, respectively. 3.4. Cd accumulation in rice plant
Fig. 1. Total Cd and DTPA-Cd in the tested soil and the proportion of DTPA-Cd to total Cd.
The Cd concentration in the rice plant of different treatments was in the order of HM > LM > ST > HN > CK z NPK (Fig. 4). The Cd
Z.-X. Rao et al. / Environmental Pollution 239 (2018) 198e204
201
Fig. 2. Total Cd and DTPA-Cd at a depth of 0e60 cm in the tested paddy soil.
Fig. 3. Total Cd and DTPA-Cd in the aggregates of the tested paddy soil. * The different lowercase letters on bars showed the significance of Cd in the same aggregates level between different treatments.
3.5. Relationship of Cd between soil and rice plant
Fig. 4. Cd in rice plants from different fertilizer treatments. * The different lowercase letters on bars showed the significance of Cd in the same parts of rice plant between different treatments.
in rice husks of NPK showed an increase of 22.0% compared with that in CK. For all parts of the rice plant, the Cd in HN treatment was 4.73%e34.8% higher than CK, with no significance. However, the Cd in the rice plant of ST treatment increased by 38.0%e98.1% compared with that in CK. In addition, the Cd in all parts of the rice plant from LM and HM was significantly increased (p < 0.01) by 59.8%e120% and 95.1%e152%, respectively, compared with CK (Fig. 4). Compared with the Cd in NPK and HN, those in the rice plants from LM and HM presented significant increases of 21.8%e 137% and 51.0%e146%, respectively (Fig. 4).
The SEM of the soil aggregates, rice plants, and fertilizer treatments exhibited a reasonable fit at c2 ¼ 0.049, df ¼ 2, p ¼ 0.976, GFI ¼ 0.999, and RMSEA ¼ 0.000. According to the path coefficient, the Cd uptake of the rice root was strongly positively affected by long-term fertilization and Cd availability in the 0.25e0.5 mm aggregates (Fig. 5). However, the effect of Cd in the 0.053e0.25 mm aggregates was weakened, and a negative direct effect and positive indirect effect on Cd uptake was observed (Table 2). For the SEM of the soil profiles, rice plants, and fertilizer treatments, the criteria for evaluating model fit were c2 ¼ 0.120, df ¼ 2, p ¼ 0.942, GFI ¼ 0.997, and RMSEA ¼ 0.000. The Cd uptake of the root was significantly affected by long-term fertilization and available Cd in the 10e20 cm soil layer (Fig. 5). The long-term impacts of different fertilization regime had significant direct effects on Cd uptake by the roots (Table 2) and indirect impacts on Cd uptake via its effect on the available Cd in the soil aggregates and profiles (Table 2). Long-term fertilization affected bulk soils, which had direct positive effects on the Cd availability in the soil aggregates. In addition, Cd accumulation in rice plants was affected by the accumulation and redistribution of available Cd in the aggregates. However, the indirect effects of fertilization on Cd uptake manifested as effects on Cd availability in the 0e10 cm soil profile, which affected Cd availability in the 10e20 cm profile directly and then had an impact on the Cd uptake of the root.
202
Z.-X. Rao et al. / Environmental Pollution 239 (2018) 198e204
Fig. 5. Final SEM framework describing the relationships among long-term fertilization, Cd availability (in aggregates and profiles), and uptake by plant.
Table 2 Direct, indirect and total effects of the different fertilization and Cd availability on the root-Cd determined via structural equation modeling (SEM). Cd in Root
SEM of aggregates
SEM of profiles
Different Fertilization Cd in Bulk Soil Cd in 1e2 mm Cd in 0.25e0.5 mm Cd in 0.053e0.25 mm Different Fertilization Cd in 0e10 cm Cd in 10e20 cm Cd in 20e30 cm
4. Discussion 4.1. Effect of long-term fertilization on Cd accumulation in surface soil As previously mentioned, the long-term application of manure fertilizers especially caused a significant accumulation of T-Cd and DTPA-Cd in the surface soil. The accumulation and availability of Cd was increasing with the usage amount of manures. A previous study also showed that a 22-year application of pig manure alone and pig manure with chemical fertilizer resulted in a significant accumulation and increasing availability of Cd in red soil (Zhou et al., 2015). Manures might contain toxic elements (including Cd) because of plant enrichment or additives in livestock feedstuff (Nicholson et al., 1999), which introduced Cd into the soil environment. For example, high contents of Cu, Zn, As, Cr, Hg, Pb and Cd were found in manure-based fertilizers from Zhejiang Province, China (Qian et al., 2016). These might indicated that manure fertilizers, rather than inorganic fertilizers, were the important causes of enhanced accumulation risks of Cd in tested paddy soil.
4.2. Effects of long-term fertilization on Cd accumulation and redistribution in soil profiles and aggregates The vertical distribution and availability of Cd was positively affected by continuous fertilization. The long-term application of high dosages of manure fertilizers caused T-Cd enrichment in 0e40 cm soil layers and relatively increase in DTPA-Cd in the 0e30 soil layers. Similarly, a significant increase of Cd in the 15e30 cm and 20e40 cm subsoil layers after 10-year and 22-year manure fertilization has been reported (Xu et al., 2015; Zhou et al., 2015).
Direct Effect
Indirect Effect
Total Effect
0.811 0 0.207 0.595 0.607 0.751 0.207 0.326 0.315
0.108 0.248 0.174 0 0.231 0.164 0.002 0.19 0
0.919 0.248 0.381 0.595 0.376 0.916 0.205 0.136 0.315
These results indicated that long-term application of manures may enhance the accumulation and availability of Cd in both topsoil and subsoil layers. The Cd distribution characteristic and grain size fraction metal loading (GSFloading, see attachment) in all the soil aggregates was also varied after long-term fertilization. It showed an increasing contribution of Cd in macro-aggregates (>0.25 mm) and a decreasing contribution in micro-aggregates of manure fertilized soils (Fig. S1). In addition, the distribution factors (DF) showed an accumulation of DTPA-Cd in medial size fraction aggregates (Fig. S1). This variation may be related to the change and redistribution of the aggregate fraction proportion which can be changed by repeated annual applications of organic fertilizers, such as cow manure, sewage sludge or municipal solid waste compost (Hojati and Nourbakhsh, 2009). However, the results from urban topsoil showed that the Cd in size fractions increased as the particle size decreased and the clay fraction of <2 mm was more seriously polluted by heavy metals than the other size fractions (Zong et al., 2016). This difference could be related to the typical drying and wetting cycles in paddy soils (Mikha et al., 2005; Gordon et al., 2008). Long-term fertilization, especially long-term application of manures affected the distribution and availability of Cd in aggregates, while further studies should be conducted.
4.3. Effects of long-term fertilization and Cd availability on Cd accumulation in rice plants The root is widely recognized as the most major connection between the rice plant and the soil environment. Meanwhile, the accumulation pattern of Cd in other parts of rice plant was similar to that in root. Thus, we only take roots into the SEM framework
Z.-X. Rao et al. / Environmental Pollution 239 (2018) 198e204
due to its vital role. The fertilization regimes had a strong impact on the available Cd in the 0e10 cm soil layer. Meanwhile, the Cd availability in the 0e10 cm profile had a direct effect on the available Cd in the 10e20 cm profile and directly affected the Cd accumulation in the roots. These findings demonstrated a pathway by which fertilization regimes impacted Cd transportation in soil profiles and then affected Cd accumulation in rice plants. The Cd uptake of the root was directly affected by the DTPA-Cd in the 10e20 cm profiles. These impacts were strongly related to the high bio-availability of Cd and abundant organic matter in the topsoil. Meanwhile, the root system of the rice plant is mainly distributed in the 0e20 cm soil layer, and less than 10% of the roots are distributed in the layers under 20 cm (Wu et al., 2011). The distribution of Cd therefore affected the uptake of root and transportation from soil into the rice plant to a large extent. The long-term impacts of different fertilizers and Cd availability in the 10e20 cm soil profiles were two main impact factors that determined the Cd accumulation in rice plants. The uptake and transportation of Cd by rice plants was also greatly positively influenced by the DTPA-Cd in the 1e2 mm and 0.25e0.5 mm aggregates of the tested paddy soil. In addition, the long-term impacts of different fertilizers had direct and indirect impacts on the Cd uptake by affecting the availability of Cd in the bulk surface soil and aggregates. Unfortunately, few papers have focused on Cd distribution in soil and rice plants considering the effects of aggregate size and fertilizer regimes. However, the present study demonstrated that the fertilizer regimes and available Cd in the 0.25e0.5 mm aggregates were indeed two critical factors affecting Cd distribution and accumulation in the ‘soil-rice plant’ system. In brief, the continuous fertilization regimes and the relative redistribution of available Cd in the soil aggregates had great influence on the Cd accumulation in rice plants. 5. Conclusions The long-term application of manure fertilizers, rather than inorganic fertilizers, led to the accumulation of Cd in surface soil and caused Cd enrichment in subsoil profiles and also led to Cd accumulation in the rice plant. The 30-year fertilization regimes resulted in redistribution of Cd in soil aggregates. The Cd accumulation in the roots was directly affected by fertilizer regimes and Cd availability in 10e20 cm soil layers and 0.25e0.5 mm aggregates. Emphasis should be placed on the application amount of manures and the original resource of Cd in manures like Cdcontained additives in feedstuff should be controlled. Acknowledgements This work was supported by the Special Fund for Agro-scientific Research in the Public Interest (201403015), the National Natural Science Foundation of China (41371318), the Australia-China Joint Research Centre e Healthy Soils for Sustainable Food Production and Environmental Quality (ACSRF48165) and the Earmarked Fund for China Agriculture Research System (CARS-16-E09). Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.envpol.2018.04.024. References Acosta, J.A., Martínez-Martínez, S., Faz, A., Arocena, J., 2011. Accumulations of major and trace elements in particle size fractions of soils on eight different parent materials. Geoderma 161, 30e42. https://doi.org/10.1016/j.geoderma.2010.12.
203
001. Arhonditsis, G.B., Stow, C.A., Steinberg, L.J., Kenney, M.A., Lathrop, R.C., McBride, S.J., Reckhow, K.H., 2006. Exploring ecological patterns with structural equation modeling and Bayesian analysis. Ecol. Modell. 192, 385e409. https://doi.org/10. 1016/j.ecolmodel.2005.07.028. Beaumelle, L., Vile, D., Lamy, I., Vandenbulcke, F., Gimbert, F., Hedde, M., 2016. A structural equation model of soil metal bioavailability to earthworms: confronting causal theory and observations using a laboratory exposure to fieldcontaminated soils. Sci. Total Environ. 569e570, 961e972. https://doi.org/10. 1016/j.scitotenv.2016.06.023. Chen, Z., Tang, Y.T., Yao, A.J., Cao, J., Wu, Z.H., Peng, Z.R., Wang, S.Z., Xiao, S., Baker, A.J.M., Qiu, R.L., 2017. Mitigation of Cd accumulation in paddy rice (Oryza sativa L.) by Fe fertilization. Environ. Pollut. 231, 549e559. https://doi.org/10. 1016/j.envpol.2017.08.055. Cheng, K., Tian, H.Z., Zhao, D., Al, E., 2013. Atmospheric emission inventory of cadmium from anthropogenic sources. Int. J. Environ. Sci. Technol. 11, 605e616. https://doi.org/10.1007/s13762-013-0206-3. Di, J., Liu, X., Du, Z., Xiao, X., Yang, G., Ren, T., 2014. Influences of long-term organic and chemical fertilization on soil aggregation and associated organic carbon fractions in a red paddy soil. Chin. J. Eco-Agriculture 22, 1129e1138. https://doi. org/10.13930/j.cnki.cjea.130121. Elliott, E.T., 1986. Aggregate structure and carbon, nitrogen, and phosphorus in native and cultivated soils. Soil Sci. Soc. Am. J. 50, 627e633. Fan, J., Ding, W., Chen, Z., Ziadi, N., 2012. Thirty-year amendment of horse manure and chemical fertilizer on the availability of micronutrients at the aggregate scale in black soil. Environ. Sci. Pollut. Res. 19, 2745e2754. https://doi.org/10. 1007/s11356-012-0774-7. Fedotov, G.N., Omel’yanyuk, G.G., Bystrova, O.N., Martynkina, E.a., Gulevskaya, V.V., Nikulina, M.V., 2008. Heavy-metal distribution in various types of soil aggregates. Dokl. Chem. 420, 125e128. https://doi.org/10.1134/S0012500808050042. Gordon, H., Haygarth, P.M., Bardgett, R.D., 2008. Drying and rewetting effects on soil microbial community composition and nutrient leaching. Soil Biol. Biochem. 40, 302e311. https://doi.org/10.1016/j.soilbio.2007.08.008. Grace, J.B., Anderson, T.M., Olff, H., Scheiner, S.M., 2010. On the specification of structural equation models for ecological systems. Ecol. Monogr. 80, 67e87. https://doi.org/10.1890/09-0464.1. Grant, C.A., Sheppard, S.C., 2008. Fertilizer impacts on cadmium availability in agricultural soils and crops. Hum. Ecol. Risk Assess. 14, 210e228. https://doi. org/10.1080/10807030801934895. , J., Schellberg, J., Srek, Hejcman, M., Szakova P., Tlustos, P., 2008. The Rengen Grassland Experiment: soil contamination by trace elements after 65 years of Ca, N, P and K fertiliser application. Nutr. Cycl. Agroecosystems 83, 39e50. https://doi.org/10.1007/s10705-008-9197-8. Hojati, S., Nourbakhsh, F., 2009. Distribution of b-glucosidase activity within aggregates of a soil amended with organic fertilizers. Am. J. Agric. Biol. Sci. 4, 179e186. https://doi.org/10.3844/ajabssp.2009.179.186. Hu, Y., Xiang, D., Veresoglou, S.D., Chen, F., Chen, Y., Hao, Z., Zhang, X., Chen, B., 2014. Soil organic carbon and soil structure are driving microbial abundance and community composition across the arid and semi-arid grasslands in northern China. Soil Biol. Biochem. 77, 51e57. https://doi.org/10.1016/j.soilbio.2014.06. 014. Ilg, K., Wilcke, W., Safronov, G., Lang, F., Fokin, A., Kaupenjohann, M., 2004. Heavy metal distribution in soil aggregates: a comparison of recent and archived aggregates from Russia. Geoderma 123, 153e162. https://doi.org/10.1016/j. geoderma.2004.02.006. Lu, R.K., 2000. Analitical Methods for Soil and Agrochemistry. Chinese agricultural Science and Technology Press. Luo, L., Ma, Y., Zhang, S., Wei, D., Zhu, Y.G., 2009. An inventory of trace element inputs to agricultural soils in China. J. Environ. Manage 90, 2524e2530. https:// doi.org/10.1016/j.jenvman.2009.01.011. Mikha, M.M., Rice, C.W., Milliken, G.A., 2005. Carbon and nitrogen mineralization as affected by drying and wetting cycles. Soil Biol. Biochem. 37, 339e347. https:// doi.org/10.1016/j.soilbio.2004.08.003. Nicholson, F.A., Chambers, B.J., Williams, J.R., Unwin, R.J., 1999. Heavy metal contents of livestock feeds and animal manures in England and Wales. Bioresour. Technol 70, 23e31. https://doi.org/10.1016/S0960-8524(99)00017-6. Qian, M., Wu, H., Wang, J., Zhang, H., Zhang, Z., Zhang, Y., Lin, H., Ma, J., 2016. Occurrence of trace elements and antibiotics in manure-based fertilizers from the Zhejiang Province of China. Sci. Total Environ. 559, 174e181. https://doi.org/ 10.1016/j.scitotenv.2016.03.123. Tang, H., Li, T., Yu, H., Zhang, X., 2016. Cadmium accumulation characteristics and removal potentials of high cadmium accumulating rice line grown in cadmiumcontaminated soils. Environ. Sci. Pollut. Res. 23, 15351e15357. https://doi.org/ 10.1007/s11356-016-6710-5. Wang, G., Zhou, L., 2017. Erratum to: application of Green manure and pig manure to Cd-Contaminated paddy soil increases the risk of Cd uptake by rice and Cd downward migration into groundwater: field micro-plot trials (2017), 228, 155, 10.1007/s11270-016-3207-2. Water. Air. Soil Pollut. 228. https://doi.org/10.1007/ s11270-017-3346-0. Wu, L., Tan, C., Liu, L., Zhu, P., Peng, C., Luo, Y., Christie, P., 2012. Cadmium bioavailability in surface soils receiving long-term applications of inorganic fertilizers and pig manure. Geoderma 173e174, 224e230. https://doi.org/10. 1016/j.geoderma.2011.12.003. Wu, W.M., Song, X.F., Sun, Z.X., Yu, Y.H., Zou, G.Y., 2011. Comparison of root distribution between different type rice. Chin. J. Rice Sci. 15, 276e280. https://doi.
204
Z.-X. Rao et al. / Environmental Pollution 239 (2018) 198e204
org/10. 16819/j. 1001 -7216. 2001. 04. 008. Xu, Y.G., Yu, W.T., Ma, Q., Zhou, H., 2015. Potential risk of cadmium in a soil-plant system as a resultof long-term (10 years) pig manure application. Plant, Soil Environ. 61, 325e375. https://doi.org/10.17221/100/2015-PSE. Zhang, F.S., Li, Y.X., Yang, M., Li, W., Yan, W.J., 2011. Cadmium concentration in animal manure and the regional pollution hazard: a case study in Northeast China. In: Bioinformatics and Biomedical Engineering, (iCBBE) 5th International Conference on Bioinformatics and Biomedical Engineering. https://doi.org/10.1109/ icbbe.2011.5781599. Zhao, F.J., Ma, Y., Zhu, Y.G., Tang, Z., McGrath, S.P., 2015. Soil contamination in China: current status and mitigation strategies. Env. Sci. Technol. 49, 750e759. https:// doi.org/10.1021/es5047099. Zhao, Y.C., Yan, Z.B., Qin, J.H., Xiao, Z.W., 2014. Effects of long-term cattle manure application on soil properties and soil heavy metals in corn seed production in
Northwest China. Env. Sci. Pollut. Res. 21, 7586e7595. https://doi.org/10.1007/ s11356-014-2671-8. Zhou, S., Liu, J., Xu, M., Lv, J., Sun, N., 2015. Accumulation, availability, and uptake of heavy metals in a red soil after 22-year fertilization and cropping. Environ. Sci. Pollut. Res. 22, 15154e15163. https://doi.org/10.1007/s11356-015-4745-7. Zhu, H., Chen, C., Xu, C., Zhu, Q., Huang, D., 2016. Effects of soil acidification and liming on the phytoavailability of cadmium in paddy soils of central subtropical China. Environ. Pollut. 219, 99e106. https://doi.org/10.1016/j.envpol.2016.10. 043. Zong, Y.T., Xiao, Q., Lu, S.G., 2016. Distribution, bioavailability, and leachability of heavy metals in soil particle size fractions of urban soils (northeastern China). Environ. Sci. Pollut. Res. 23, 14600e14607. https://doi.org/10.1007/s11356-0166652-y.
Contents lists available at ScienceDirect
Environmental Pollution journal homepage: www.elsevier.com/locate/envpol
Distribution and availability of cadmium in profile and aggregates of a paddy soil with 30-year fertilization and its impact on Cd accumulation in rice plant* Zhong-Xiu Rao a, b, Dao-You Huang a, Jin-Shui Wu a, Qi-Hong Zhu a, *, Han-Hua Zhu a, Chao Xu a, Jie Xiong a, b, Hui Wang a, Ming-Meng Duan a, b a
Key Laboratory for Agro-ecological Processes in Subtropical Region, Institute of Subtropical Agriculture, Chinese Academy of Sciences, Changsha, Hunan, 410125, China University of Chinese Academy of Sciences, Beijing, 100049, China
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 17 November 2017 Received in revised form 5 April 2018 Accepted 5 April 2018
The research was conducted to investigate the accumulation, distribution and availability of Cd in paddy soil and their relation to Cd in rice plants under 30-year fertilization regimes. Six treatments were involved in the study: control without fertilization (CK), chemical fertilizer (NPK), high nitrogen chemical fertilizer (HN), rice straw incorporation (ST), low and high dosage of manure fertilizer (LM and HM). Total and DTPA extractable concentration of Cd (T-Cd and DTPA-Cd) in bulk soils (20 cm topsoil), profiles (0 e60 cm) and aggregates (>2, 1e2, 0.5e1, 0.25e0.5, 0.053e0.25 and < 0.053 mm) were investigated. The Cd concentration in relevant rice plant (roots, stems, leaves, husks and grains) were also analyzed. Manure fertilizers caused T-Cd accumulation in bulk soil with a significant increase of 36.2% in LM and 81.2% in HM. Similar impacts of manure fertilizers were observed in DTPA-Cd in the bulk soil. Further, the HM generated a further accumulation in deeper soil layers, presenting a remarkable increase of T-Cd (28.3%e225%) in 10e40 cm and DTPA-Cd (116%e158%) in 10e30 cm profiles. Moreover, the continuous application of manure fertilizers enhanced the availability of Cd in all aggregate size classes with an increase of 17.3%e87.8% in DTPA-Cd. Organic fertilizers (LM, HM and ST) heightened the content of Cd (38.0%e152%) in all parts of rice plant. The accumulation of Cd in rice plants was directly affected by fertilization regimes and Cd availability in the 10e20 cm soil layers and 0.25e0.5 mm aggregates. In conclusion, long-term application of manures resulted in increasing availability of Cd in aggregates and in topsoil and subsoil layers, which accordingly enhanced the accumulation of Cd in rice plants. © 2018 Elsevier Ltd. All rights reserved.
Keywords: Paddy soil Aggregates Profile Rice Cadmium (Cd) Accumulation
1. Introduction Cadmium (Cd) is a hazardous heavy metal found in the environment. Anthropogenic activities have resulted in the release of a substantial amount of Cd into cropland and especially in paddy regions (Grant and Sheppard, 2008). Considering the high transfer ability of Cd in soil-rice system, the contamination risk of Cd in rice grain has been found all over the world, e.g. China and Southeast Asia (Zhao et al., 2015; Tang et al., 2016). Controlling the accumulation of Cd in rice grains have been a worldwide issue because of
*
This paper has been recommended for acceptance by B. Nowack. * Corresponding author. E-mail address: [email protected] (Q.-H. Zhu).
https://doi.org/10.1016/j.envpol.2018.04.024 0269-7491/© 2018 Elsevier Ltd. All rights reserved.
the bioaccumulation of Cd in crops and its potential risks to human body through the food chain (Xu et al., 2015; Chen et al., 2017). Previous studies indicated that Cd-bearing fertilizers, especially manures, may be one of the important sources for Cd entering into soil (Zhang et al., 2011; Cheng et al., 2013; Zhao et al., 2014). For example, livestock manures were observed to account for 55% of the total Cd inputs in agricultural soils of China (Luo et al., 2009). The effect of fertilization on Cd in soil has been related not only to the concentration and availability of Cd in bulk soils but also to the accumulation and distribution of Cd in soil profiles and aggregates (Acosta et al., 2011; Fan et al., 2012). Application of pig manures has been reported to significantly increase the Cd concentration in soils (Wu et al., 2012; Xu et al., 2015). It was also observed that Cd in subsoil layers increased significantly after continuous application of pig manures for 10 years (Xu et al., 2015). Furthermore, the
Z.-X. Rao et al. / Environmental Pollution 239 (2018) 198e204
availability of Cd may be strongly affected by soil aggregate size distribution because aggregates are the basic unit of soil construction and important place for chemical and biological reaction activities (Ilg et al., 2004; Fedotov et al., 2008). Increasing risk of Cd uptake by rice was also found in paddy soils fertilized with pig manure and green manure (Wang and Zhou, 2017). Long-term fertilization is likely to result in much more serious accumulation risks of heavy metals in soils and crops (Hejcman et al., 2008; Zhao et al., 2015; Zhou et al., 2015). It is crucial to determine the distribution of Cd, and its availability, at the profile and aggregate scales, as affected by long-term fertilization regimes. Therefore, the distribution and accumulation characteristics and availability of Cd in bulk paddy soil, profiles, aggregates and rice plants were analyzed using a 30-year fertilization experiment. The objective was to elucidate the effects of long-term fertilization on Cd availability in soil profiles and aggregates and relate their impacts on Cd accumulation in rice plant. 2. Materials and methods 2.1. Site and soil description The long-term fertilization experiment was initiated in 1986 in Hunan Province, China. The site has a humid subtropical monsoon climate, an annual average temperature of 16.8 C and a mean precipitation of 1358 mm. The tested soil was derived from modern river alluvial deposits and classified as alluvial sand soil. The pH of the tested soil is 5.6 on average, and it contains an average quality of soil organic carbon (SOC) 21.1 g kg1, total nitrogen (TN) 2.17 g kg1, total phosphorus (TP) 1.08 g kg1 and total potassium (TK) 22.3 g kg1. 2.2. Fertilization designs The fertilization regimes had been applied and lasted for 30 years up to 2015, in which year the soil samples were collected. The fertilization treatments were: no fertilizer (CK), chemical fertilizer (NPK), high nitrogen chemical fertilizer (HN), rice straw incorporation (ST), low dosage of manure fertilizer (LM) and high dosage of manure fertilizer (HM). The dosage of chemical fertilizer (NPK) was determined according to the N, P and K supplying capacity of the soil and the absorption ability of plants in the supervised field. The source of N and K were urea fertilizer (46%) and potassium chloride fertilizer (60%), respectively. The P was applied in the form of calcium-magnesia phosphate fertilizer (12%). Thus, some other elements such as Ca, Mg and Si were also imported into soil coupled with P. High nitrogen chemical fertilizer (HN) involved further application of N, additional to the NPK treatment. The manure fertilizer used in the study was mainly pig and fowl manure, and the proportions of N in the low and high dosages of manure fertilizer were 30% and 60% of total N applied to the soil, respectively. The amount of ST was 3000 kg ha1. The LM, HM, and ST treatments were supplemented by chemical fertilizers. The details of fertilizer application are shown in Table 1. 2.3. Sampling The bulk soil samples were collected from the 0e20 cm surface soil layer. To ensure the representation and uniformity of soil samples, a composite soil sample was obtained by thoroughly mixing 5 subsamples at each sampling site. Bulk soil samples were pre-treated by air drying and then ground to pass through 1 mm and 0.149 mm sieves for further analysis. Profile samples were collected in the soil layers from depths of 0e60 cm and were divided into 6 layers with 10 cm per layer. A
199
Table 1 Field fertilization amount of tested soils (by purity nutrient content, kg/ha/a) (Di et al., 2014). Treatments
Chemical fertilizer
Straw/Manure
Total amount
N
P
K
N
P
K
N
P
K
NPK HN ST LM HM
457 757 378 313 164
44 44 36 30 36
160 160 1 61 0
0 0 81 147 293
0 0 12 56 111
0 0 167 106 211
457 757 459 460 457
44 44 48 86 147
160 160 168 167 211
*Field fertilization amount (purity content) represented the average values between 2004 and 2011.
stainless steel earth drill with a scale line mark was used for sample collection. The profile samples were air dried and sieved to <0.149 mm for further analysis. Water-stable aggregates were derived from 0 to 20 cm topsoil and separated into six size fractions by wet sieving (Elliott, 1986) as, >2, 1e2, 0.5e1, 0.25e0.5, 0.053e0.25 and < 0.053 mm. Five sequential size sieves were used and placed in descending order from top to bottom. Before fractionation, the soil samples were submersed in distilled water for 10 min and then sieved vertically in distilled water by machine at a vertical reciprocating frequency of 35 r min1 over a period of 15 min. After separation, the samples in the sieve were collected into aluminum specimen boxes, and the samples <0.053 mm in the water were collected by settlement and centrifugation. All aggregate samples were then dried at 60e70 C in an oven at constant temperature. The plant samples were also collected together with the soil samples. The rice plants were collected in their entirety. After deactivation of enzymes at 105 C, the rice plants were then divided into roots, stems, leaves, husks and grains. All the plant sample components were dried in oven at 60e70 C and then ground into powder with a stainless steel pulverizer for further analysis. 2.4. Physical and chemical analysis The soil physical and chemical properties (SOC, pH, and total N/ P/K) were analyzed according to Lu (2000). The total Cd in the bulk soil, water-stable aggregates and soil profiles was solubilized by digestion of 0.5 g in 10 mL aqua regia (HCl: HNO3 ¼ 3:1, V/V; AR) and 2 mL HClO4 (AR). The available Cd was extracted with a DTPATEA-CaCl2 solution by shaking for 2 h at 180 r min1. This DTPA method is widely used and regarded as national standard protocols to investigate soil extractable Cd in China (Zhu et al., 2016). The Cd in rice plants was solubilized by digestion of 1.0 g in 10 mL HNO3 (AR) and 2 mL HClO4 (AR). All detected soil and plant samples were digested or extracted in two duplicate. Distilled deionized water was used for blank correction and the referenced soil (GBW07405) was used as certified reference material. Both of the blank and reference material were digested in the same pattern of soil samples. All digestion and extraction solutions were finally analyzed via inductively coupled plasma optical emission spectrometry (ICPOES, Varian, US), with a limit of 0.003 mg mL1 for Cd. A stock solution of Cd (1000 mg mL1, Merck) was used as an external standard by diluting it into 1 mg mL1. Moreover, all data were blankcorrected and the standard curve was checked every ten samples during the detection procedure. 2.5. Statistical analysis All statistical analyses were performed via Excel 2010 and SPSS
200
Z.-X. Rao et al. / Environmental Pollution 239 (2018) 198e204
19.0 statistical software (SPSS Inc., Chicago, IL, US). Statistically significant differences between the means were tested by an ANOVA within a general linear model (GLM). The least significant difference (LSD) was used to assess the significance at the 95% confidence level. Structural equation modeling (SEM) was performed via Amos 24.0 software. The structural equation modeling (SEM) is a suitable method to provide insight into the relationships of physical, chemical and biological variables in ecological research (Arhonditsis et al., 2006; Grace et al., 2010). It focuses on understanding direct and indirect pathways, to extract scientific understanding about systems (Grace et al., 2010). In the present study, the SEM framework was applied to investigate the direct and indirect effects of Cd bioavailability in soil profiles and aggregates on Cd in rice plants. All data were standardized to achieve linearity and normality. A maximum likelihood method was used to estimate the parameters. The goodnessof-fit of the model was evaluated by a Satarra-Bentler scaled chisquare statistic. The model was rejected if the p-value of the c2 test was below the significance level 0.05. Several variables in the final structure model were excluded because their paths were nonsignificant and inconsistent with the c2 and p-value of the model. The model can be re-specified according to modification indices. A given path can then be added if the modification index is above 4 and scientifically supported (Beaumelle et al., 2016). Several criteria, such as the goodness-of-fit index (GFI), comparative fit index (CFI) and root mean square error of approximation (RMSEA), were used for evaluation of model fit (Hu et al., 2014; Beaumelle et al., 2016). Good model fits were reflected by RMSEA values below 0.05 and GFI values above 0.90. 3. Results 3.1. Accumulation and availability of Cd in bulk soil The total Cd (T-Cd) and DTPA-extracted Cd (DTPA-Cd) in the bulk soil ranged from 0.33 to 0.72 mg kg1 and 0.17e0.36 mg kg1, respectively (Fig. 1). The T-Cd in the control (CK) was 0.40 mg kg1, which was higher than that of the Chinese soil environmental quality standards (Class Ⅱ, 0.3 mg kg1). Compared with CK, the TCd decreased significantly (p < 0.05) by 16.5% in NPK, whereas no significant differences were observed in HN and ST. However, the TCd in HM and LM increased significantly (p < 0.01) by 81.2% and 36.2% compared with that in CK, respectively, and by 116% and 63.1% compared with that in NPK, respectively (Fig. 1). The DTPA-Cd in CK was 0.21 mg kg1, with a proportion of 53.0% of T-Cd. Compared with CK, the DTPA-Cd in HM and LM increased
significantly (p < 0.01) by 73.6% and 28.5%. However, the DTPA-Cd in NPK was significantly lower (18.2%) than that in CK. No significant differences were observed between HN and ST treatments. Compared with the DTPA-Cd in NPK, those in the HM, LM and ST treatments were significantly increased (p < 0.01) by 112%, 57.1% and 24.3%, respectively (Fig. 1). The proportion of DTPA-Cd in T-Cd of all fertilizer treatments was 50.0%e59.5%, with the highest proportion in the ST. 3.2. Distribution and availability of Cd in the soil profiles The T-Cd in the soil profiles of CK was relatively high in the 0e20 cm surface layer and then sharply declined in the profiles below 30 cm, slipped to a range of 0.14e0.19 mg kg1. Compared with CK, the HM had the highest concentration of T-Cd, which showed a significant increase (p < 0.01) of 28.3%e225% in 0e40 cm soil profiles (Fig. 2). In addition, the T-Cd in HM of 0e40 cm profiles was significantly (p < 0.01) 45.7%e395% precede those in NPK, HN and ST (Fig. 2). The T-Cd in 0e10 cm soil layer of LM was significantly higher (p < 0.01) by 23.1% than that in CK (Fig. 2). The DTPA-Cd in CK declined as the depth of the soil layers increased, with little change observed in 30e60 cm soil profiles (Fig. 2). Similarly, the DTPA-Cd in the other fertilized treatments had the same pattern as that of the CK, and the concentration tended to be uniform in 40e60 cm soil profiles. In addition, no significant differences in 20e60 cm profiles were observed for CK, NPK, HN, ST and LM. However, the DTPA-Cd in 0e10 cm profile of the LM and 0e30 cm profiles of the HM increased significantly (p < 0.01) by 48.7% and by 69.3%e157%, respectively, compared with that in the CK (Fig. 2). 3.3. Distribution and availability of Cd in soil aggregates The T-Cd in the aggregates of the CK ranged from 0.27 to 0.80 mg kg1, with the highest concentration in the >2 mm aggregates and the lowest concentration in the <0.053 mm aggregates. Compared to CK, the T-Cd in the aggregates of LM significantly increased by 4.35%e48.6% (p < 0.01) in 0.5e1 mm and 0.053e0.25 mm aggregates. In addition, the T-Cd in the aggregates (except for 0.5e1 mm) of HM (p < 0.01) increased significantly by 13.9%e70.0% (Fig. 3). The T-Cd in >2 mm and 0.5e1 mm aggregates of ST, HN and NPK showed a significant decrease of 12.3%e60.6% compared with that in the CK, whereas an increase (significantly) of 19.6%e77.6% was observed in the <0.053 mm aggregates (Fig. 3). When compared with the T-Cd in NPK and HN, those in the >2 mm aggregates in ST, LM and HM increased gradually and significantly by 14.9%e285% (Fig. 3). Similarly, the T-Cd in the 1e2 mm aggregates increased by 15.8%e41.9% in ST, LM, and LM compared with those in NPK and HN (Fig. 3). The DTPA-Cd content ranged from 0.11 to 0.26 mg kg1 and was relatively high in the 0.25e0.5 mm aggregates followed by the 0.5e1 mm aggregates in the CK. Compared with the DTPA-Cd in the CK, those in LM and HM increased significantly (p < 0.01) by 17.3%e 87.7% and 41.1%e75.9% in all aggregates, respectively (Fig. 3). However, the DTPA-Cd in ST only showed a significant decrease of 21.0% in >2 mm aggregates. Similarly, significant results were found in the 0.25e0.5 mm and 0.053e0.25 mm aggregates of NPK and >2 mm and 0.5e1 mm aggregates of HN (Fig. 3). Compared to the inorganic treatments, DTPA-Cd in the LM and HM increased by 8.97%e153% and 34.5%e124% (p < 0.01) in aggregates, respectively. 3.4. Cd accumulation in rice plant
Fig. 1. Total Cd and DTPA-Cd in the tested soil and the proportion of DTPA-Cd to total Cd.
The Cd concentration in the rice plant of different treatments was in the order of HM > LM > ST > HN > CK z NPK (Fig. 4). The Cd
Z.-X. Rao et al. / Environmental Pollution 239 (2018) 198e204
201
Fig. 2. Total Cd and DTPA-Cd at a depth of 0e60 cm in the tested paddy soil.
Fig. 3. Total Cd and DTPA-Cd in the aggregates of the tested paddy soil. * The different lowercase letters on bars showed the significance of Cd in the same aggregates level between different treatments.
3.5. Relationship of Cd between soil and rice plant
Fig. 4. Cd in rice plants from different fertilizer treatments. * The different lowercase letters on bars showed the significance of Cd in the same parts of rice plant between different treatments.
in rice husks of NPK showed an increase of 22.0% compared with that in CK. For all parts of the rice plant, the Cd in HN treatment was 4.73%e34.8% higher than CK, with no significance. However, the Cd in the rice plant of ST treatment increased by 38.0%e98.1% compared with that in CK. In addition, the Cd in all parts of the rice plant from LM and HM was significantly increased (p < 0.01) by 59.8%e120% and 95.1%e152%, respectively, compared with CK (Fig. 4). Compared with the Cd in NPK and HN, those in the rice plants from LM and HM presented significant increases of 21.8%e 137% and 51.0%e146%, respectively (Fig. 4).
The SEM of the soil aggregates, rice plants, and fertilizer treatments exhibited a reasonable fit at c2 ¼ 0.049, df ¼ 2, p ¼ 0.976, GFI ¼ 0.999, and RMSEA ¼ 0.000. According to the path coefficient, the Cd uptake of the rice root was strongly positively affected by long-term fertilization and Cd availability in the 0.25e0.5 mm aggregates (Fig. 5). However, the effect of Cd in the 0.053e0.25 mm aggregates was weakened, and a negative direct effect and positive indirect effect on Cd uptake was observed (Table 2). For the SEM of the soil profiles, rice plants, and fertilizer treatments, the criteria for evaluating model fit were c2 ¼ 0.120, df ¼ 2, p ¼ 0.942, GFI ¼ 0.997, and RMSEA ¼ 0.000. The Cd uptake of the root was significantly affected by long-term fertilization and available Cd in the 10e20 cm soil layer (Fig. 5). The long-term impacts of different fertilization regime had significant direct effects on Cd uptake by the roots (Table 2) and indirect impacts on Cd uptake via its effect on the available Cd in the soil aggregates and profiles (Table 2). Long-term fertilization affected bulk soils, which had direct positive effects on the Cd availability in the soil aggregates. In addition, Cd accumulation in rice plants was affected by the accumulation and redistribution of available Cd in the aggregates. However, the indirect effects of fertilization on Cd uptake manifested as effects on Cd availability in the 0e10 cm soil profile, which affected Cd availability in the 10e20 cm profile directly and then had an impact on the Cd uptake of the root.
202
Z.-X. Rao et al. / Environmental Pollution 239 (2018) 198e204
Fig. 5. Final SEM framework describing the relationships among long-term fertilization, Cd availability (in aggregates and profiles), and uptake by plant.
Table 2 Direct, indirect and total effects of the different fertilization and Cd availability on the root-Cd determined via structural equation modeling (SEM). Cd in Root
SEM of aggregates
SEM of profiles
Different Fertilization Cd in Bulk Soil Cd in 1e2 mm Cd in 0.25e0.5 mm Cd in 0.053e0.25 mm Different Fertilization Cd in 0e10 cm Cd in 10e20 cm Cd in 20e30 cm
4. Discussion 4.1. Effect of long-term fertilization on Cd accumulation in surface soil As previously mentioned, the long-term application of manure fertilizers especially caused a significant accumulation of T-Cd and DTPA-Cd in the surface soil. The accumulation and availability of Cd was increasing with the usage amount of manures. A previous study also showed that a 22-year application of pig manure alone and pig manure with chemical fertilizer resulted in a significant accumulation and increasing availability of Cd in red soil (Zhou et al., 2015). Manures might contain toxic elements (including Cd) because of plant enrichment or additives in livestock feedstuff (Nicholson et al., 1999), which introduced Cd into the soil environment. For example, high contents of Cu, Zn, As, Cr, Hg, Pb and Cd were found in manure-based fertilizers from Zhejiang Province, China (Qian et al., 2016). These might indicated that manure fertilizers, rather than inorganic fertilizers, were the important causes of enhanced accumulation risks of Cd in tested paddy soil.
4.2. Effects of long-term fertilization on Cd accumulation and redistribution in soil profiles and aggregates The vertical distribution and availability of Cd was positively affected by continuous fertilization. The long-term application of high dosages of manure fertilizers caused T-Cd enrichment in 0e40 cm soil layers and relatively increase in DTPA-Cd in the 0e30 soil layers. Similarly, a significant increase of Cd in the 15e30 cm and 20e40 cm subsoil layers after 10-year and 22-year manure fertilization has been reported (Xu et al., 2015; Zhou et al., 2015).
Direct Effect
Indirect Effect
Total Effect
0.811 0 0.207 0.595 0.607 0.751 0.207 0.326 0.315
0.108 0.248 0.174 0 0.231 0.164 0.002 0.19 0
0.919 0.248 0.381 0.595 0.376 0.916 0.205 0.136 0.315
These results indicated that long-term application of manures may enhance the accumulation and availability of Cd in both topsoil and subsoil layers. The Cd distribution characteristic and grain size fraction metal loading (GSFloading, see attachment) in all the soil aggregates was also varied after long-term fertilization. It showed an increasing contribution of Cd in macro-aggregates (>0.25 mm) and a decreasing contribution in micro-aggregates of manure fertilized soils (Fig. S1). In addition, the distribution factors (DF) showed an accumulation of DTPA-Cd in medial size fraction aggregates (Fig. S1). This variation may be related to the change and redistribution of the aggregate fraction proportion which can be changed by repeated annual applications of organic fertilizers, such as cow manure, sewage sludge or municipal solid waste compost (Hojati and Nourbakhsh, 2009). However, the results from urban topsoil showed that the Cd in size fractions increased as the particle size decreased and the clay fraction of <2 mm was more seriously polluted by heavy metals than the other size fractions (Zong et al., 2016). This difference could be related to the typical drying and wetting cycles in paddy soils (Mikha et al., 2005; Gordon et al., 2008). Long-term fertilization, especially long-term application of manures affected the distribution and availability of Cd in aggregates, while further studies should be conducted.
4.3. Effects of long-term fertilization and Cd availability on Cd accumulation in rice plants The root is widely recognized as the most major connection between the rice plant and the soil environment. Meanwhile, the accumulation pattern of Cd in other parts of rice plant was similar to that in root. Thus, we only take roots into the SEM framework
Z.-X. Rao et al. / Environmental Pollution 239 (2018) 198e204
due to its vital role. The fertilization regimes had a strong impact on the available Cd in the 0e10 cm soil layer. Meanwhile, the Cd availability in the 0e10 cm profile had a direct effect on the available Cd in the 10e20 cm profile and directly affected the Cd accumulation in the roots. These findings demonstrated a pathway by which fertilization regimes impacted Cd transportation in soil profiles and then affected Cd accumulation in rice plants. The Cd uptake of the root was directly affected by the DTPA-Cd in the 10e20 cm profiles. These impacts were strongly related to the high bio-availability of Cd and abundant organic matter in the topsoil. Meanwhile, the root system of the rice plant is mainly distributed in the 0e20 cm soil layer, and less than 10% of the roots are distributed in the layers under 20 cm (Wu et al., 2011). The distribution of Cd therefore affected the uptake of root and transportation from soil into the rice plant to a large extent. The long-term impacts of different fertilizers and Cd availability in the 10e20 cm soil profiles were two main impact factors that determined the Cd accumulation in rice plants. The uptake and transportation of Cd by rice plants was also greatly positively influenced by the DTPA-Cd in the 1e2 mm and 0.25e0.5 mm aggregates of the tested paddy soil. In addition, the long-term impacts of different fertilizers had direct and indirect impacts on the Cd uptake by affecting the availability of Cd in the bulk surface soil and aggregates. Unfortunately, few papers have focused on Cd distribution in soil and rice plants considering the effects of aggregate size and fertilizer regimes. However, the present study demonstrated that the fertilizer regimes and available Cd in the 0.25e0.5 mm aggregates were indeed two critical factors affecting Cd distribution and accumulation in the ‘soil-rice plant’ system. In brief, the continuous fertilization regimes and the relative redistribution of available Cd in the soil aggregates had great influence on the Cd accumulation in rice plants. 5. Conclusions The long-term application of manure fertilizers, rather than inorganic fertilizers, led to the accumulation of Cd in surface soil and caused Cd enrichment in subsoil profiles and also led to Cd accumulation in the rice plant. The 30-year fertilization regimes resulted in redistribution of Cd in soil aggregates. The Cd accumulation in the roots was directly affected by fertilizer regimes and Cd availability in 10e20 cm soil layers and 0.25e0.5 mm aggregates. Emphasis should be placed on the application amount of manures and the original resource of Cd in manures like Cdcontained additives in feedstuff should be controlled. Acknowledgements This work was supported by the Special Fund for Agro-scientific Research in the Public Interest (201403015), the National Natural Science Foundation of China (41371318), the Australia-China Joint Research Centre e Healthy Soils for Sustainable Food Production and Environmental Quality (ACSRF48165) and the Earmarked Fund for China Agriculture Research System (CARS-16-E09). Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.envpol.2018.04.024. References Acosta, J.A., Martínez-Martínez, S., Faz, A., Arocena, J., 2011. Accumulations of major and trace elements in particle size fractions of soils on eight different parent materials. Geoderma 161, 30e42. https://doi.org/10.1016/j.geoderma.2010.12.
203
001. Arhonditsis, G.B., Stow, C.A., Steinberg, L.J., Kenney, M.A., Lathrop, R.C., McBride, S.J., Reckhow, K.H., 2006. Exploring ecological patterns with structural equation modeling and Bayesian analysis. Ecol. Modell. 192, 385e409. https://doi.org/10. 1016/j.ecolmodel.2005.07.028. Beaumelle, L., Vile, D., Lamy, I., Vandenbulcke, F., Gimbert, F., Hedde, M., 2016. A structural equation model of soil metal bioavailability to earthworms: confronting causal theory and observations using a laboratory exposure to fieldcontaminated soils. Sci. Total Environ. 569e570, 961e972. https://doi.org/10. 1016/j.scitotenv.2016.06.023. Chen, Z., Tang, Y.T., Yao, A.J., Cao, J., Wu, Z.H., Peng, Z.R., Wang, S.Z., Xiao, S., Baker, A.J.M., Qiu, R.L., 2017. Mitigation of Cd accumulation in paddy rice (Oryza sativa L.) by Fe fertilization. Environ. Pollut. 231, 549e559. https://doi.org/10. 1016/j.envpol.2017.08.055. Cheng, K., Tian, H.Z., Zhao, D., Al, E., 2013. Atmospheric emission inventory of cadmium from anthropogenic sources. Int. J. Environ. Sci. Technol. 11, 605e616. https://doi.org/10.1007/s13762-013-0206-3. Di, J., Liu, X., Du, Z., Xiao, X., Yang, G., Ren, T., 2014. Influences of long-term organic and chemical fertilization on soil aggregation and associated organic carbon fractions in a red paddy soil. Chin. J. Eco-Agriculture 22, 1129e1138. https://doi. org/10.13930/j.cnki.cjea.130121. Elliott, E.T., 1986. Aggregate structure and carbon, nitrogen, and phosphorus in native and cultivated soils. Soil Sci. Soc. Am. J. 50, 627e633. Fan, J., Ding, W., Chen, Z., Ziadi, N., 2012. Thirty-year amendment of horse manure and chemical fertilizer on the availability of micronutrients at the aggregate scale in black soil. Environ. Sci. Pollut. Res. 19, 2745e2754. https://doi.org/10. 1007/s11356-012-0774-7. Fedotov, G.N., Omel’yanyuk, G.G., Bystrova, O.N., Martynkina, E.a., Gulevskaya, V.V., Nikulina, M.V., 2008. Heavy-metal distribution in various types of soil aggregates. Dokl. Chem. 420, 125e128. https://doi.org/10.1134/S0012500808050042. Gordon, H., Haygarth, P.M., Bardgett, R.D., 2008. Drying and rewetting effects on soil microbial community composition and nutrient leaching. Soil Biol. Biochem. 40, 302e311. https://doi.org/10.1016/j.soilbio.2007.08.008. Grace, J.B., Anderson, T.M., Olff, H., Scheiner, S.M., 2010. On the specification of structural equation models for ecological systems. Ecol. Monogr. 80, 67e87. https://doi.org/10.1890/09-0464.1. Grant, C.A., Sheppard, S.C., 2008. Fertilizer impacts on cadmium availability in agricultural soils and crops. Hum. Ecol. Risk Assess. 14, 210e228. https://doi. org/10.1080/10807030801934895. , J., Schellberg, J., Srek, Hejcman, M., Szakova P., Tlustos, P., 2008. The Rengen Grassland Experiment: soil contamination by trace elements after 65 years of Ca, N, P and K fertiliser application. Nutr. Cycl. Agroecosystems 83, 39e50. https://doi.org/10.1007/s10705-008-9197-8. Hojati, S., Nourbakhsh, F., 2009. Distribution of b-glucosidase activity within aggregates of a soil amended with organic fertilizers. Am. J. Agric. Biol. Sci. 4, 179e186. https://doi.org/10.3844/ajabssp.2009.179.186. Hu, Y., Xiang, D., Veresoglou, S.D., Chen, F., Chen, Y., Hao, Z., Zhang, X., Chen, B., 2014. Soil organic carbon and soil structure are driving microbial abundance and community composition across the arid and semi-arid grasslands in northern China. Soil Biol. Biochem. 77, 51e57. https://doi.org/10.1016/j.soilbio.2014.06. 014. Ilg, K., Wilcke, W., Safronov, G., Lang, F., Fokin, A., Kaupenjohann, M., 2004. Heavy metal distribution in soil aggregates: a comparison of recent and archived aggregates from Russia. Geoderma 123, 153e162. https://doi.org/10.1016/j. geoderma.2004.02.006. Lu, R.K., 2000. Analitical Methods for Soil and Agrochemistry. Chinese agricultural Science and Technology Press. Luo, L., Ma, Y., Zhang, S., Wei, D., Zhu, Y.G., 2009. An inventory of trace element inputs to agricultural soils in China. J. Environ. Manage 90, 2524e2530. https:// doi.org/10.1016/j.jenvman.2009.01.011. Mikha, M.M., Rice, C.W., Milliken, G.A., 2005. Carbon and nitrogen mineralization as affected by drying and wetting cycles. Soil Biol. Biochem. 37, 339e347. https:// doi.org/10.1016/j.soilbio.2004.08.003. Nicholson, F.A., Chambers, B.J., Williams, J.R., Unwin, R.J., 1999. Heavy metal contents of livestock feeds and animal manures in England and Wales. Bioresour. Technol 70, 23e31. https://doi.org/10.1016/S0960-8524(99)00017-6. Qian, M., Wu, H., Wang, J., Zhang, H., Zhang, Z., Zhang, Y., Lin, H., Ma, J., 2016. Occurrence of trace elements and antibiotics in manure-based fertilizers from the Zhejiang Province of China. Sci. Total Environ. 559, 174e181. https://doi.org/ 10.1016/j.scitotenv.2016.03.123. Tang, H., Li, T., Yu, H., Zhang, X., 2016. Cadmium accumulation characteristics and removal potentials of high cadmium accumulating rice line grown in cadmiumcontaminated soils. Environ. Sci. Pollut. Res. 23, 15351e15357. https://doi.org/ 10.1007/s11356-016-6710-5. Wang, G., Zhou, L., 2017. Erratum to: application of Green manure and pig manure to Cd-Contaminated paddy soil increases the risk of Cd uptake by rice and Cd downward migration into groundwater: field micro-plot trials (2017), 228, 155, 10.1007/s11270-016-3207-2. Water. Air. Soil Pollut. 228. https://doi.org/10.1007/ s11270-017-3346-0. Wu, L., Tan, C., Liu, L., Zhu, P., Peng, C., Luo, Y., Christie, P., 2012. Cadmium bioavailability in surface soils receiving long-term applications of inorganic fertilizers and pig manure. Geoderma 173e174, 224e230. https://doi.org/10. 1016/j.geoderma.2011.12.003. Wu, W.M., Song, X.F., Sun, Z.X., Yu, Y.H., Zou, G.Y., 2011. Comparison of root distribution between different type rice. Chin. J. Rice Sci. 15, 276e280. https://doi.
204
Z.-X. Rao et al. / Environmental Pollution 239 (2018) 198e204
org/10. 16819/j. 1001 -7216. 2001. 04. 008. Xu, Y.G., Yu, W.T., Ma, Q., Zhou, H., 2015. Potential risk of cadmium in a soil-plant system as a resultof long-term (10 years) pig manure application. Plant, Soil Environ. 61, 325e375. https://doi.org/10.17221/100/2015-PSE. Zhang, F.S., Li, Y.X., Yang, M., Li, W., Yan, W.J., 2011. Cadmium concentration in animal manure and the regional pollution hazard: a case study in Northeast China. In: Bioinformatics and Biomedical Engineering, (iCBBE) 5th International Conference on Bioinformatics and Biomedical Engineering. https://doi.org/10.1109/ icbbe.2011.5781599. Zhao, F.J., Ma, Y., Zhu, Y.G., Tang, Z., McGrath, S.P., 2015. Soil contamination in China: current status and mitigation strategies. Env. Sci. Technol. 49, 750e759. https:// doi.org/10.1021/es5047099. Zhao, Y.C., Yan, Z.B., Qin, J.H., Xiao, Z.W., 2014. Effects of long-term cattle manure application on soil properties and soil heavy metals in corn seed production in
Northwest China. Env. Sci. Pollut. Res. 21, 7586e7595. https://doi.org/10.1007/ s11356-014-2671-8. Zhou, S., Liu, J., Xu, M., Lv, J., Sun, N., 2015. Accumulation, availability, and uptake of heavy metals in a red soil after 22-year fertilization and cropping. Environ. Sci. Pollut. Res. 22, 15154e15163. https://doi.org/10.1007/s11356-015-4745-7. Zhu, H., Chen, C., Xu, C., Zhu, Q., Huang, D., 2016. Effects of soil acidification and liming on the phytoavailability of cadmium in paddy soils of central subtropical China. Environ. Pollut. 219, 99e106. https://doi.org/10.1016/j.envpol.2016.10. 043. Zong, Y.T., Xiao, Q., Lu, S.G., 2016. Distribution, bioavailability, and leachability of heavy metals in soil particle size fractions of urban soils (northeastern China). Environ. Sci. Pollut. Res. 23, 14600e14607. https://doi.org/10.1007/s11356-0166652-y.